Hein G Gooszen

Introduction: Results of the first randomized trial comparing on-demand versus planned-relaparotomy strategy in patients with severe peritonitis (RELAP trial) indicated no clear differences in primary outcomes. We now report the full economic evaluation for this trial, including detailed methods, nonmedical costs, further differentiated cost calculations, and robustness of different assumptions in sensitivity analyses. Methods: An economic evaluation was conducted from a societal perspective alongside a randomized controlled trial in 229 patients with severe secondary peritonitis and an acute physiology and chronic health evaluation (APACHE)-II score ≥11 from two academic and five regional teaching hospitals in the Netherlands. After the index laparotomy, patients were randomly allocated to an on-demand or a planned-relaparotomy strategy. Primary resource-utilization data were used to estimate mean total costs per patient during the index admission and after discharge until 1 year after the index operation. Overall differences in costs between the on-demand relaparotomy strategy and the planned strategy, as well as relative differences across several clinical subgroups, were evaluated. Results: Costs were substantially lower in the on-demand group (mean, €65,768 versus €83,450 per patient in the planned group; mean absolute difference, €17,682; 95% CI, €5,062 to €29,004). Relative differences in mean total costs per patient (approximately 21%) were robust to various alternative assumptions. Planned relaparotomy consistently generated more costs across the whole range of different courses of disease (quick recovery and few resources used on one end of the spectrum; slow recovery and many resources used on the other end). This difference in costs between the two surgical strategies also did not vary significantly across several clinical subgroups. Conclusions: The reduction in societal costs renders the on-demand strategy a more-efficient relaparotomy strategy in patients with severe peritonitis. These differences were found across the full range of healthcare resources as well as across patients with different courses of disease. Trial Registration: ISRCTN51729393

Background: Laparoscopic Nissen fundoplication (LNF) has essentially replaced its conventional open counterpart (CNF). An economic evaluation of LNF compared with CNF based on prospective data with adequate follow-up is lacking. Methods: Data from two consecutive studies (a randomized clinical trial (RCT) of 57 patients undergoing LNF and 46 undergoing CNF that was terminated prematurely, and a follow-up study of 121 consecutive patients with LNF) were combined to determine incremental cost-effectiveness 1 year after surgery. Results: Mean operating time, reoperation rate and hospital costs of LNF were lower in the second series. The mean overall hospital cost per patient was €9126 for LNF and €6989 for CNF at 1 year in the initial RCT, and €7782 in the second LNF series. The success rate of both LNF and CNF at 1 year was 91 per cent in the RCT, and LNF was successful in 90·1 per cent in the second series. A cost reduction of €998 for LNF would cancel out the cost advantage of CNF. Similarly, if the reoperation rate after LNF decreased from 0·05 to below 0·008 and/or if the mean duration of sick leave after LNF was reduced from 67·2 to less than 61·1 days, the procedure would become less expensive than CNF. Complications, reoperation rate and quality of life after both operations were similar. Conclusion: Including reinterventions, the outcome at 1 year after LNF and CNF was similar. In a well organized setting with appropriate expertise, the cost advantage of CNF may be neutralized.

Background: Infectious complications are the major cause of death in acute pancreatitis. Small bowel bacterial overgrowth and subsequent bacterial translocation are held responsible for the vast majority of these infections. Goal of this study is to determine whether selected probiotics are capable of preventing infectious complications without the disadvantages of antibiotic prophylaxis; antibiotic resistance and fungal overgrowth. Methods/design: PROPATRIA is a double-blind, placebo-controlled randomised multicenter trial in which 200 patients will be randomly allocated to a multispecies probiotic preparation (Ecologic 641) or placebo. The study is performed in all 8 Dutch University Hospitals and 7 non-University hospitals. The study-product is administered twice daily through a nasojejunal tube for 28 days or until discharge. Patients eligible for randomisation are adult patients with a first onset of predicted severe acute pancreatitis: Imrie criteria 3 or more, CRP 150 mg/L or more, APACHE II score 8 or more. Exclusion criteria are post-ERCP pancreatitis, malignancy, infection/sepsis caused by a second disease, intra-operative diagnosis of pancreatitis and use of probiotics during the study. Administration of the study product is started within 72 hours after onset of abdominal pain. The primary endpoint is the total number of infectious complications. Secondary endpoints are mortality, necrosectomy, antibiotic resistance, hospital stay and adverse events. To demonstrate that probiotic prophylaxis reduces the proportion of patients with infectious complications from 50% to 30%, with alpha 0,05 and power 80%, a total sample size of 200 patients was calculated. Conclusion: The PROPATRIA study is aimed to show a reduction in infectious complications due to early enteral use of multispecies probiotics in severe acute pancreatitis.

Background: This study assessed the long-term effect of Nissen fundoplication on oesophageal and oesophagogastric junction (OGJ) motility. Methods: Symptoms were scored and oesophageal manometry performed in 34 consecutive patients with chronic gastro-oesophageal reflux disease, before, 3 months after and 2 years after surgery. Results: Distal peristaltic amplitude increased from a median of 57 (95 per cent confidence interval (c.i.) 18 to 107) mmHg to 86 (95 per cent c.i. 54 to 208) mmHg (P < 0·001) at 3 months and 92 (45 to 210) mmHg (P < 0·001) at 2 years. In four patients the amplitude increased to more than 180 mmHg and three of these patients reported odynophagia. After surgery, a linear relationship was observed between the peristaltic amplitude and nadir OGJ relaxation pressure at 3 months (rs = 0·68, P < 0·001) and 2 years (rs = 0·64, P < 0·001). A significant correlation was also found between amplitude and both basal OGJ pressure and intrabolus pressure at 3 months (rs = 0·58, P < 0·001 and rs = 0·63, P < 0·001 respectively) and 2 years (rs = 0·71, P < 0·001 and rs = 0·49, P = 0·024). There was a relationship between peristaltic amplitude and the odynophagia score at 2 years (rs = 0·60, P = 0·017). Conclusion: Within 3 months of fundoplication the amplitude of oesophageal peristalsis increased substantially, leading to a nutcracker oesophagus and odynophagia in a subgroup of patients. These phenomena did not appear to progress with time.

The acceptance for publication and the implementation of the results of a randomised clinical trial are determined not only by the quality of the study, but also by its timing and the prevailing opinion prior to the study. In a randomised study, comparing laparoscopic and open fundoplication in patients with gastro-oesophageal reflux, interim analysis showed statistically significantly more complications related to laparoscopic intervention. Publication of the study led to an avalanche of criticism focusing mainly on the inexperience of the surgeons. The cost-effectiveness analysis, showing that open fundoplication had fewer complications and was cheaper, was repeatedly rejected by established journals mainly because of criticism of surgical aspects of the study. In the meanwhile a study on a second group of 100 patients was initiated. They underwent laparoscopic fundoplication conducted by two experienced surgeons. The results were statistically significantly better than in the laparoscopic arm of the initial randomised trial. The learning curve proved to be longer than had previously been thought.

Background The importance of anatomical reposition and fixation of the type I hiatal hernia during antireflux surgery has often been emphasized. It is not known whether the initial anatomical repair withstands the test of time and whether this repair is necessary for a successful outcome. Methods The relationship between the objective anatomical and subjective symptomatic outcome of Nissen fundoplication was investigated prospectively in 57 patients. Findings of herniation, telescoping and obstruction at the level of the lower oesophageal sphincter on barium swallow were scored 2 years after operation by investigators who were unaware of the symptoms, and were related to symptoms and patient satisfaction evaluated by a standard questionnaire. Results According to strict criteria, some 55 per cent of patients had some degree of anatomical failure; if only complete herniation, significant telescoping and signs of obstruction were scored as abnormal, 27 per cent had anatomical failure. There was no relation to subjective outcome; relief was reported by 48 of 49 patients, 25 of whom were cured and 23 significantly improved. Conclusion Anatomical repair during antireflux surgery does not stand the test of time. Although this has no demonstrable influence on the subjective outcome, the authors do not recommend deviating from well designed surgical guidelines. Current theories on the mechanism of antireflux surgery require further evaluation.

Background: Duodenum-preserving resection of the head of the pancreas (DPRHP) according to Beger has been developed as an alternative to pylorus-preserving resection of the pancreatic head for painful chronic pancreatitis. Methods: Between 1988 and 2000, 36 consecutive DPRHPs were performed. The group was divided into patients with (group 1; n = 23) and without (group 2; n = 13) significant enlargement of the pancreatic head. Pain was the indication for surgery in all patients. Results: Complications occurred in 12 patients, necessitating reoperation in 11. Initial overall results were favourable; significant improvement or complete relief of pain was reported in 27 of 35 patients. Long-term results were obtained in 27 of 30 patients; the overall success rate was 16 of 27, 13 of 16 patients with distinct enlargement of the pancreatic head and 3 of the 11 with minimal or no enlargement (P = 0·018). Conclusion: DPRHP can be performed with good early results. This effect is sustained in patients with distinct localized disease of the pancreatic head. In those without, the long-term results are disappointing.

Background: After Nissen fundoplication, troublesome dysphagia develops in 5–10 per cent of patients. The mechanism of dysphagia has not been fully resolved, in spite of a number of studies focusing on oesophageal motility and lower oesophageal sphincter (LOS) dynamics. Tightness and length of the wrap have had considerable attention, without giving a fully satisfactory explanation of the pathophysiological mechanism. Methods: Eighteen patients with persistent dysphagia after Nissen fundoplication needing reoperation were studied. Eighteen patients, matched for age and sex, without dysphagia after Nissen fundoplication were used as controls. Reoperation consisted of conversion of a 360° into a 270° wrap. Barium swallow, endoscopy, oesophageal manometry and 24-h pH monitoring were performed before and after (re)operation. Results: Peristaltic amplitude, velocity and duration of contraction were not significantly influenced by operation. In 16 of 18 patients with dysphagia, LOS relaxation was incomplete and the residual relaxation pressure was significantly higher than that in the group without dysphagia (P < 0·01). No correlation was found between LOS pressure and peristaltic amplitude, nor between LOS pressure and ramp pressure in the distal oesophagus. After reoperation, basal LOS pressure decreased significantly (P < 0·01) and LOS relaxation was complete in all but three patients; residual relaxation pressure decreased (P < 0·01) and was significantly lower than that after uncomplicated Nissen fundoplication. In the latter group, LOS pressure, residual relaxation pressure and ramp pressure increased significantly after operation (P < 0·01). Conclusion: A return to complete LOS relaxation and a decrease in residual relaxation pressure play an important role in resolving dysphagia.