Monitoring caustic injuries from emergency department databases using automatic keyword recognition software.

In Italy the European Union Injury Database reports the involvement of chemical products in 0.9% of home and leisure accidents. The Emergency Department registry on domestic accidents in Italy and the Poison Control Centres record that 90% of cases of exposure to toxic substances occur in the home. It is not rare for the effects of chemical agents to be observed in hospitals, with a high potential risk of damage - the rate of this cause of hospital admission is double the domestic injury average. The aim of this study was to monitor the effects of injuries caused by caustic agents in Italy using automatic free-text recognition in Emergency Department medical databases. We created a Stata software program to automatically identify caustic or corrosive injury cases using an agent-specific list of keywords. We focused attention on the procedure's sensitivity and specificity. Ten hospitals in six regions of Italy participated in the study. The program identified 112 cases of injury by caustic or corrosive agents. Checking the cases by quality controls (based on manual reading of ED reports), we assessed 99 cases as true positive, i.e. 88.4% of the patients were automatically recognized by the software as being affected by caustic substances (99% CI: 80.6%- 96.2%), that is to say 0.59% (99% CI: 0.45%-0.76%) of the whole sample of home injuries, a value almost three times as high as that expected (p < 0.0001) from European codified information. False positives were 11.6% of the recognized cases (99% CI: 5.1%- 21.5%). Our automatic procedure for caustic agent identification proved to have excellent product recognition capacity with an acceptable level of excess sensitivity. Contrary to our a priori hypothesis, the automatic recognition system provided a level of identification of agents possessing caustic effects that was significantly much greater than was predictable on the basis of the values from current codifications reported in the European Database.

Road safety: objective of the European Union for 2010 and assessment criteria.

A few years ago the European Union proposed that its member states take steps to reduce the number of road accidents and their health and social consequences (number deceased, injured, handicapped, etc.). The main objective of this ongoing challenge is a 50% reduction of mortality by these events by 2010. The European Union closely monitors, also through dedicated workshops, projects and achievements of member states. In our opinion, however, too much attention seems to be dedicated to a reduction in the percentage of mortality. It undoubtedly merits attention, but lacking other elements, there is the danger of providing data, in terms of assessment and communication, which may lead to a distortion of the facts. This paper clarifies our point of view and shows why the reduction in the percentage of mortality per accident as well as the starting mortality index should be considered in order to provide a more accurate picture of the trend in the various countries working towards this objective.

[Prevalence estimate of three dichotomic characteristics by the analysis of pools, with an application for simulation of the joint detection of cocaine, cannabis and alcohol]. / Stima della prevalenza di tre caratteristiche dicotomiche tramite analisi di pool, con un'applicazione per simulazione alla ricerca congiunta di cocaina, cannabis ed alcol.

In this paper a method is introduced for estimating from analytical results obtained from pools of individual samples, the prevalences of three substances and their associations. The treated data are dichotomic, i.e. presence-absence of every substance in the analyzed pool. In fact, this method is the extension to three characteristics of the PERM (Politomic Erased Respondent Method), of which the previous version considered only two characteristics. Besides the theoretical development of the method, a simulated application to checks for driving under the influence of alcohol, cocaine and cannabis, is reported. The general agreement among the estimates obtained from the analytical results of the pools and those obtained from the results of the individual samples, is close (within few percent points). As expected this agreement improves when the number of analyzed pools increases. The proposed method could find an interesting application in the checks related to road safety, but also in those regarding work safety and sport safety.

[Impact of increasing number of road alcohol controls on the prevalence of driving under the influence in Italy]. / L'aumento del numero dei controlli alcolemici su strada e i suoi effetti sulla prevalenza della guida in stato di ebbrezza in Italia.

In Italy the number of road controls for the driving under the influence of alcohol is significantly grown in the last years. In 2006 these controls have been around 250,000; in 2007, around 800,000. On the grounds of the data currently available one can estimate around 1,500,000 controls in 2008. At the same time a lowering percentage of drivers with blood alcohol concentration greather than the legal limit (equal to 0.5 g/l in Italy) has been recorded. Aforesaid percentage amounted to 15% and 6% in 2006 and 2007 respectively, with a lessening of 9 percent points. This reduction could plausibly be the outcome of the increased controls number on the italian roads. Likely, increased perception to be checked by the police has induced to a greater caution the drivers. However the controls are not random controls, but based on a 'well-founded suspicion', that is when the police is right to hold by their behavior that the drivers can be in state of drunkenness. This procedure prevents us from properly estimating the true effect. Therefore, carry out a suitable number of random controls besides the well-founded controls seems needed to quantify in accurate statistical way the proportion of drivers in state of drunkenness as well as its temporal trend.

The urgency of establishing a rapid monitoring system for mortality due to traffic accidents (as well as for all mortality due to violence and accidents).

The data regarding the number of deaths from road accidents per year in Italy is presently available only after extensive time, in the range of years. This situation originates from problems related to data collection as well as to the subsequent management of the information. The two sources which allow the quantification of this mortality are: statistic of "Road Accidents documented by the Police Force", consisting in accidents reported by the Police and involving deceased or injured victims; general mortality statistics The first source presents problems related to quality and completeness; the second, problems related to dimensions: reports of mortality due to traffic accidents (presently about 6000 reports per year) end up, as they should, in the flow of general mortality (more than 560,000 reports per year), which necessarily lengthens response time. This excessive time interval between notification of deaths from traffic accidents and their actual availability in terms of statistics has, as of today, a negative impact not only on the prompt knowledge of the phenomenon, but also on the necessary appraisal in light of reaching the objectives proposed by the European Union by 2010 (reduction of 50% health consequences of traffic accidents). This article aims to show that it would be easy to set up a Rapid Monitoring System for deaths due to traffic accidents by creating an appropriate sub-flow of the general mortality statistics. Based on reported appraisals, we aim to show that it seems highly practicable (and convenient) to extend this system to all accidental and violent mortality, inclusive of drug-related mortality (presently underestimated). The authors feel that the System herein described could, in a matter of months, provide, at very low costs, useful and reliable information regarding health and social issues. It is worthwhile pointing out that what hereby proposed does not interfere in any way with the flow of general mortality currently in use and may also be used for parallel collection of data which is currently not being recorded.

[Measuring blood alcohol concentration on the road as a tool for early identification of subjects with alcohol-related problems]. / I controlli alcolemici su strada quale strumento per l'identificazione precoce di soggetti con problemi alcol-correlati.

This paper deals with some considerations showing the phenomenon 'alcohol and driving' doesn't have primary character. Instead, it springs from the individual's state leading him to drink over and over again too much alcoholic beverages. The fact that such a subject will drive in state of drunkenness afterwards is accidental, even if frequent. Therefore, to improve the road safety, as well as the social safety, action on the individual as regards its excessive alcohol consumption is needed. Nevertheless, for different reasons (bad information, underestimation of the risk, social reproval), who has problems with alcohol tends to hide himself; therefore problematic subjects are not easy to be identified. To get round this impediment, we propose to use BAC (Blood Alcohol Concentration) drivers' controls on the road to identify such as subjects, taking as cutoff a BAC greater than 1.5 g/l, or at least twice overcoming the legal limit (0.5 g/l), apart from the BAC value in itself In this way it would be possible to offer an early chance of specialistic assistance beforehand to help them to settle their own problem with the alcohol. This approach could be a successful strategy since among the subjects that would come so identified many are clearly in acceptable psycophysical state (as they can drive); many are young people (therefore subjects at the beginning of drinker's career); practically all have a good grounds for solving their problem (to get driving licence back). Of course, if this program had to have put into effect it would be necessary to strengthen the specialised Alcohol Services and therefore to bear the cost of it. Nevertheless, the outcomes would have significant spin-off not only on the road safety but, in more broad terms, on the social safety too.

[Drivers' random controls and population's recent drug use monitoring]. / I controlli casuali dei conducenti di veicoli a motore quale base per il monitoraggio dell'uso recente di droghe nella popolazione.

Substance abuse, particularly among young people, does not seem to have the tendency to decrease. Our knowledges on this phenomenon are manifold and they validly compete to address the actions of contrast. Nevertheless, it would seem profit to be able to have further informations, to place side by side to those already existing, with the aim to improve the surveillance of the phenomenon. For this purpose, in the present paper we propose a monitoring model based on the results of random controls on road, carried out by the Police (or by the Hospital) in relationship to the road safety. The representativeness of the data that we could get this way appears elevated as an hight percentage of the population own the driver's licence. As it is shown, these controls could be both individual and related to drivers' pools of biological samples. This last approach would seem to be more practicable since problems relating to the drivers' privacy would be avoided.

[A method to estimate one's own blood alcohol concentration when the ministerial tables are not avaible]. / Un metodo per stimare la propria alcolemia quando non si disponga delle tabelle ministeriali.

To reduce the prevalence of driving under the influence, tables allowing to estimate one's own blood alcohol concentration (BAC) by type and quantity of alcoholic drinks intake have been enacted by decree in Italy. Such tables, based on a modified Widmark's formula, are now put up in all public concerns serving alcoholic beverages. Aim of this initiative is to try to get subjects which consume alcoholics and then will drive a vehicle take in account their own estimated BAC and, on this base, put into effect, if needed, suitable actions (to avoid or to limit a further consumption, to wait more time before driving, to leave driving a sober subject). Nevertheless, many occasions exist in which these tables are not available. To allow anybody to rough estimate his own BAC in these cases too, a proper method has been developed. Briefly, the weight (in grams) of consumed alcohol has to be divided by half her own weight, if female drunk on an empty stomach (by the 90% of her own weight, if she drunk on a full stomach); by 70% of his own weight, if male drunk on an empty stomach (by 120% of his own weight, if he drunk in a full stomach). Consistency between BAC values estimated by the proposed method and those shown in the ministerial tables is very narrow: they differ in a few hundredth of grams/liter. Unlike the ministerial tables, the proposed method needs to compute the grams of ingested alcohol. This maybe involves some difficulties that, nevertheless, can be overcome easily. In our opinion, the skillfulness in computing the grams of assumed alcohol is of great significance since it provides the subject with a strong signal not only in road safety terms, but also in health terms. The ministerial tables and the proposed method should be part of teaching to issue the driving licence and to recovery of driving licence taken away points. In broad terms, the school should teach youngs to calculate alcohol quantities assumed by drink to acquaint them with the risks paving the way for a more aware drinking when they will come age.

Systematic review: Endoscopic dilatation in Crohn's disease.

BACKGROUND: Endoscopic dilatation for Crohn's disease has been evaluated only in some small and heterogeneous studies. AIM: To evaluate any association between the main clinical variables and endoscopic variables and the efficacy and safety of endoscopic dilatation in Crohn's disease. METHODS: A Medline search regarding pneumatic dilatation in Crohn's disease was performed. Several technical and clinical variables were extracted from each study to build up a descriptive, pool-data analysis. Data on individual patients were extracted from suitable studies to create a simulated population upon which a multivariate statistical analysis was performed. RESULTS: Thirteen studies enrolling 347 Crohn's disease patients were reviewed. Endoscopic dilatation was mainly applied to postsurgical strictures, being technically successful in 86% of the cases. Long-term clinical efficacy was achieved in 58% of the patients. Mean follow-up was as long as 33 months, corresponding to 800 patient years of follow-up. Major complication rate was 2%, being higher than 10% in two series. At multivariate analysis, a stricture length < or = 4 cm was associated with a surgery-free outcome (OR: 4.01; 95% CI: 1.16-13.8; P < 0.028). CONCLUSIONS: Endoscopic dilatation is an effective and safe treatment for short strictures caused by Crohn's disease, impacting substantially on the natural history of these patients.

Quantifying a phenomenon without knowledge of individual data: the erased respondent method (ERM).

A new method (ERM-Erased Respondent Method) is presented that deletes every link between the respondent subject and the sensitive responses he provides. The method is alternative to Randomized Response Technique (RRT) by S.L. Warner (1965), moreover it is applicable also to dosages of biological sample pools. The ERM is based on the binomial distribution where, in place of the single trial (single individual) probability of success (positive result) is taken into account the probability of failure (negative result) of a group of n individuals. Applying the method to N groups, each one including n individuals, it is possible to calculate the proportion of "positive" individuals among the examined subjects (point estimation of the population proportion) by means of the proportion of "negative" groups observed. This method, which could also be used in studies concerning biological samples (forming N pools each one consisting of biological materials from n subjects), appears to be of interest for its applicability in phenomena of difficult observation, whereby the subjects are little inclined to provide the requested response or whereby linking the subject to his responses (or analytical results) could entail undesired consequences for him, even of ethical or legal liability. In conclusion, the method allows knowing certain characteristics of a population without emphasizing individual conditions. In this way, it appears probable that the ERM can be applicable to problems of confidentiality of the information, such as sexual abuse, exit-polls, some genetic diseases, doping agents, alcohol abuse or psychotropic substances addiction, especially linked to vehicles driving. In any case is to be expected that ERM method will be used especially for the analysis of biologic material pools, where the subject is not involved in any way in producing the results and is totally absent any linkage to him.

Road safety and the tsunami of cell phones.

In the last years an extremely rapid massive diffusion of the cell phones is occurred. Currently in Italy, as in many other countries, almost all the youngsters and a lot of adults possess a cell phones. This device is enormously useful, but its use can determine negative effects on the user's attention, i.e. during vehicle driving. Numerous epidemiological studies show that cell phone use in driving determines a relative risk of causing a road accident around 4 (equivalent to driving with 0.8 g/l blood alcohol concentration), for both hands-held and hands-free devices. This risk doesn't seem well perceived from the Italian drivers: as our surveys show, the 2.5% of them use a cellular hands-held while driving. This situation appears to evolve toward more critical conditions, given the quick diffusion of ever more technologically advanced instruments, such as video-cell phones, mostly capable of capturing the attention, or TV-cell phones, with which the distracting occasions could be more frequent and prolonged. The use of the cell phone seems to imply an important risk for pedestrians: the first results of the monitoring results that we are producing on this matter in Rome show that the 5.5% of the pedestrians cross the road while talking with a hands-held cell phones, in the most of cases ignoring the traffic conditions at all. All this facts show the urgency to promote specific actions of prevention, in absence of which it is reasonable to foresee a consistent growth in the number of road traffic accidents. In our opinion it is necessary to capillarily inform the public of the signalled risks, possibly also in the advertising spaces of the producers and managers of the cellular telephony, apart obvious repressive actions on the use of the cell phones during the guide of a vehicle.

Colon cancer prevention in Italy: cost-effectiveness analysis with CT colonography and endoscopy.

BACKGROUND: Colorectal cancer (CRC) is a major cause of mortality in Italy. Although prevention of CRC is possible, its cost-effectiveness when applied to the Italian population is unknown. Recently, computerized tomographic colonography (CTC) has been proposed for CRC screening. AIM: To compare the efficacy and cost-effectiveness of CTC screening in a simulated Italian population with those of colonoscopy and flexible sigmoidoscopy (FS). METHODS: The cost-effectiveness of different screening strategies was compared using a Markov process computer model, in which in a hypothetical population of 100,000 50 year-olds were investigated by CTC, colonoscopy or FS every decade. Outcomes were projected to the Italian national level. RESULTS: CRC incidence reduction was calculated at 40.9%, 38.2%, and 31.8% with colonoscopy, CTC and FS, respectively. As compared to no screening, all screening programs were shown to be cost-saving, allowing a saving of 11 Euro, 17 Euro, and 48 Euro per person with colonoscopy, FS and CTC, respectively. FS appeared to be less cost-effective than CTC, whilst colonoscopy appeared to be an expensive option as compared to CTC. Undiscounted national expenditure was calculated to be 1,042,489,512 Euro, 1,093,268,285 Euro, and 1,198,783,428 Euro for FS, CTC and colonoscopy, respectively, as compared to 695,818,078 Euro without screening. CONCLUSION: CRC screening is cost-saving in Italy, irrespective of the technique applied. CTC appeared to be more cost-effective than FS, and it may also become a valid alternative to colonoscopy.

Politomic Erased Respondent Method (PERM) for the joint assessment of two dichotomous characteristics, with an application to the polydrug use of cocaine and cannabis by simulation.

The recently introduced ERM (Erased Respondent Method) allows the prevalence of a characteristic to be estimated in a certain population by using the results obtained from groups of subjects rather than the data from single subjects. In the case of qualitative analytical surveys, the method allows the said prevalence to be estimated on the basis of the results derived from the analyses of pools of single samples. In this way, the knowledge of the individual data is not longer needed for the purpose of the estimates. This is particularly useful when there is sensitive data. In this work, it is shown how the ERM can also be applied to polytomic-type responses (e.g. negative, positive to A, positive to B, positive to A and B) in relation to qualitative analytical surveys. It is therefore possible to also estimate the prevalence of positive subjects to more characteristics on the basis of this extension of the ERM, we named it PERM (Politomic ERM). An application of the method is given by simulation in this work: an hypothetical street check in which one can estimate the prevalence of drivers under the influence of cocaine and/or cannabis (estimate of the prevalence of the users of a single substance and the prevalence of polydrug users). An important application of the method seems up to now to be in the casual street monitoring of the driver of a vehicle under the influence of alcohol and/or drugs of abuse.

Quantitative Erased Respondent Method (QERM): using quantitative group data as an alternative to quantitative individual data. A simulated application to road checks for drivers under the influence of cocaine.

The Erased Respondent Method (ERM) allows the prevalence of a characteristic to be estimated without needing to know the state of the individual statistical units. A recent extension of the method, which also did not make use of individual data, also estimated the prevalence of two or more characteristics in the same individual (PERM, Politomic ERM). In this paper, ERM is extended further to quantitative data coming from analytical determinations, considering pools of samples instead of individual samples. The extension of the method above described, we named it with the acronym QERM (Quantitative ERM), allows to estimate (only on the ground of analytical determinations made from pools of individual samples) the average and the standard deviation of the distribution of the determinations concerning only positive individual samples. By referring to road controls for driving under the influence of abuse drugs that constitute pools of biological material from more drivers, with the ERM we are able to estimate the prevalence of subjects that are positive for a certain substance, while with the QERM we can estimate the mean and standard deviation of the quantity of the substance measured in biological liquids only from the subjects that are positive to the substance. In order to better clarify this new method, an application to a road check for driving under the influence of cocaine will also be presented that was obtained by simulation. The application fields in Public Health for ERM, PERM and QERM are many; however in our opinion the most interesting areas of application are road traffic, occupational and sport safety.

[Comparison among ecography, scintigraphy and surgery in the localization of parathyroid glands in uremic patients undergoing parathyroidectomy]. / Confronto tra ecografia, scintigrafia e chirurgia nella localiz- zazione delle ghiandole paratiroidi nei pazienti uremici in dialisi sottoposti a paratiroidectomia.

BACKGROUND: Sensitivity and specificity of the most widely employed techniques of parathyroid glands localization, namely echography and scintigraphy, are mostly obtained with short-term follow-up data and do not underline the existence of a methodological problem. As a matter of fact, both methods identify only pathological glands, with no "normal" results; therefore "true negatives" cannot be obtained. Aim of our study was to compare, by means of a statistically appropriate approach, the results of echography, scintigraphy and surgery with the data obtained after a mid term follow-up period, enabling us to discover all parathyroid glands. METHODS: Twenty six consecutive dialysis patients (14M/12F; age 50+/-12 years) underwent echography and scintigraphy immediately before a total parathyroidectomy with autotransplantation and were followed-up for 6 months to recognize all the existing glands (PTH levels and scintigraphy). RESULTS: Total identified glands were: 73 by scintigraphy, 86 by echography, 99 by surgery and 103 by follow-up data. The concordance indexes (K0) between the number of glands effectively present in the individual patient (follow-up data) and those identified with each method were rather low with scintigraphy (0.071) and echography (0.218), and acceptable (0.578) with surgery. The number of patients correctly classified was: 9/26 (34,6%) with scintigraphy, 13/26 (50%) with echography and 22/26 (85%) with surgery. Finally, the number of wrongly identified glands (from zero to three) in each patient was similar with scintigraphy (65,4%) and echography (50%) and significantly better with surgery (15,6%; p<0.01). CONCLUSIONS: The most reliable technique to identify parathyroid glands in uremic subjects is surgery, nonetheless a meticulous clinical follow-up is necessary to recognize all of them.

Early clinical predictive factors during coma recovery.

In severe brain injury patients few studies have examined the role of early clinical factors emerging before recovery of consciousness. Patients suffering from vegetative state and minimally conscious state in fact may need variable periods of time for recovery of the ability to follow commands. In a previous study we retrospectively examined a population of very severe traumatic brain injury patients with coma duration of at least 15 days (prolonged coma), and we found, as significant predictive factors for the final outcome, the time interval from brain injury to the recovery of the following clinical variables: optical fixation, spontaneous motor activity and first safe oral feeding. Psychomotor agitation and bulimia during coma recovery were also favourable prognostic factors for the final outcome. In a further study, also as for the neuropsychological recovery, the clinical variable with the best significant predictive value was the interval from head trauma to the recovery of safe oral feeding. In the present study the presence of psychomotor agitation diagnosed by means of LCF (score 4 = confused-agitated) at the admission time in rehabilitation predicted a statistically significant better outcome at the discharge time in comparison with patients without agitation.

Effect of Italy's motorcycle helmet law on traumatic brain injuries.

OBJECTIVES: To evaluate the impact of a revised Italian motorcycle-moped-scooter helmet law on crash brain injuries. DESIGN: A pre-post law evaluation of helmet use and traumatic brain injury (TBI) occurrence from 1999 to 2001. SETTING: Romagna region, northeastern Italy, with a 2000 resident population of 983 534 persons. PARTICIPANTS: Motorcycle-moped rider survey for helmet use compliance and all residents in the region admitted to the Division of Neurosurgery of the Maurizio Bufalini Hospital in Cesena, Italy for TBI. OUTCOME MEASURES: Helmet use compliance and change in TBI admissions and type(s) of brain lesions. RESULTS: Helmet use increased from an average of less than 20% to over 96%. A comparison of TBI incidence in the Romagna region shows that there was no significant variation before and after introduction of the revised helmet law, except for TBI admissions for motorcycle-moped crashes where a 66% decrease was observed. In the same area TBI admissions by age group showed that motorcycle mopeds riders aged 14-60 years sustained significantly fewer TBIs. The rate of TBI admissions to neurosurgery decreased by over 31% and epidural hematomas almost completely disappeared in crash injured moped riders. CONCLUSIONS: The revised Italian mandatory helmet law, with police enforcement, is an effective measure for TBI prevention at all ages.

Gastric cancer prevalence in patients with liver cirrhosis.

An increased risk for gastric cancer in patients with liver cirrhosis has recently been reported in epidemiological studies. The present endoscopic study was performed to further evaluate whether people with cirrhosis are at increased risk for gastric cancer development. We reviewed the medical records of all cirrhotic patients referred to our Endoscopic Service for portal hypertension screening and, therefore, cases of latent gastric cancer were observed. For a comparison, the prevalence (age and sex standardized) of latent gastric cancer in the general population was estimated hypothesizing a latency period of 5 years. Overall, 1379 patients with cirrhosis were selected from a total of 15 791 endoscopically examined different patients observed during the period 1982-1997. Histological assessment revealed the presence of gastric cancer in 10 patients (9 males and 1 female). There was a significant 2.6-fold (P<0.01) increase in prevalence of gastric cancer compared with that expected in our cirrhotic patients. In conclusion, our findings confirm that liver cirrhosis would seem to be a risk factor for the development of gastric cancer. Other studies are needed to evaluate the pathogenic mechanisms involved.

Long-term outcome of endoscopic pneumatic dilatation in Crohn's disease.

BACKGROUND: To avoid multiple surgeries in stenosing Crohn's disease, pneumatic endoscopic dilatation has been introduced. The present study evaluated the long-term clinical outcome in Crohn's disease patients after endoscopic dilatation for ileal or neoileal strictures. PATIENTS AND METHODS: All Crohn's disease patients who underwent pneumatic dilatation of ileal or ileo-colonic strictures between January 1988 and December 2001 were invited to return for a clinical check-up in June 2002. Clinical, endoscopic and radiological reports were reviewed. Symptomatic relief from sub-occlusive symptoms without requiring surgery was considered as a positive outcome, whereas the requirement of surgery was regarded as an unfavourable outcome. Possible predictors of favourable outcome were analysed. RESULTS: Endoscopic dilatation was technically successful in 34/43 (79%) Crohn's disease patients, with a mean number of dilatations per patient of 3 +/- 3.13. During a mean follow-up of 63.7 +/- 44.6 months, a positive long-term outcome was observed in 18 (52.9%) patients, whereas surgery was necessary in the remaining 16 cases. The risk of surgery was distinctly higher within 2 years post-dilatation than in the next 2 years (26.4% versus 8.3%, respectively; P = 0.078). No clear clinical, endoscopic or radiological predictive factors for a successful outcome were identified. CONCLUSIONS: Endoscopic pneumatic dilatation is an effective and safe procedure to be applied to patients with stenosing Crohn's disease, offering a very long-term benefit in a sub-group of patients.