Upper endoscopy cannot be proposed as a routinary investigation, it needs to be performed in the case of alarm symptoms (haematmesis, melena or ulcer-like acitve clinical signs, which occur simultaneously to NSAID administration). Moreover, it may investigate only gastro-duodenal injuries and not intestinal ones which are more frequent than estimated. A specific histological picture does not exist for NSAID damage, it may be useful to detect a simultaneous presence of Helicobacter pylori. Dyspeptic symptoms may be related to a lot of factors and are not specific. On the light of cost-benefit ratio, I think that a good approach could be the detection of the blood in the stools and fecal calprotectin baseline and in the course of treatment. Chenges in the results may indicate a potential iatrogen effect and stimulate further investigations in a limted sample of patients.

Is the patient symptomatic? or is it that the patient has been on NSAIDs for 4 weeks and you just want to evaluate? I was not clear from your statement if you are saying you have a patient on NSAIDs who is having dyspeptic symptoms - what are you defining as dyspeptic symptoms? Is your patient having abdominal pain, nausea, vomiting, bloating/belching? Do they have alarm symptoms i.e hematemesis/hematochezia? how old is the patient?

I should confirm the above mentioned comments and like also to know the exact setting you are asking the question. Are you planning a research proposal or are you asking to plan for your daily practice?

Answering the 2nd option is easier. No guideline exists that recommends on evaluating GI tract before a short course NSAID use with invasive techniques like endoscopy. So relying on history taking remains as the sole tool available. One should remember however that the GI damage by NSAIDs are usually asymptomatic. Complications like bleeding can occur without any previous alarming symptom!

But if you are planning for a research, another approach and design is needed. So what is your main aim of the question?

This would be for research. I need to evaluate the damage caused by 2 different NSAIDs during a 4-week tratment.

I think that you can use a questionnaire for symptom onset, baseline and final detection of blood and calprotectin in the stools and evaluate the need of endoscopy for alarm symptoms (I think that the short period of therapy will not require this kind of investigation). Another parameter you may evaluate before therapy is the presence of H. pylori, which could be a bias for your study. And, finally you may evaluate the need of anti-acids or PPIs.

Dear Thais

Now the setting is clear for me. I think you should think of the known complications of the NSAIDs on all the GI tract. We all know that the most common side effect of these group of drugs is on the upper GU tract. But we should also remember that there are some less common complications in the middle & lower GI tract. These complications like ulcers in small intestine and webs and colitis and ulcers in colon are seen with long-term use of NSAIDs and I do not think that a 4 wks course of any NSAID will cause a significant problem in those areas; so in this regard checking the history of the patients and asking for diarrhea, abdominal pain and rectal bleeding along with stool OB and calprotectin as Dr Enzo recommended is all that is needed.

But for the upper GI tract as you are going to have a research I think apart from the questionnaire you should do an upper endoscopy in the beginning & at the end of 4 weeks course of the treatment period. As this is not a routine recommendation and you will need to do this procedure, you will need an informed consent to get a permission from the subjects. If you go non-invasive and do not do gastroscopy, you will miss a lot of gastric and duodenal erosions, which are usually asymptomatic and they will not make any change in stool OB etc.

I agree with Iradj Maleki about the opporunity to perform an upper endoscopy in order to detect mucosal erosions free of symptoms even if the cost/benefit must be evauated for this study. However, lesions by NSAIDs in small bowel are strongly understimated because of the difficulty to investigate this tract. So at least indirect markers of occult/obscure bleeding and inflammation as well as the exclusion of H. pylori positive and assuming PPI subjects are very important to design the study. I should like to remember that PPIs may exert a protective effect on upper tract, but they are strongly suspected to worsen lower tract lesions for their ability to change pH and consequently microbiota composition.

Endoscopy for NSAID-induced damage is the typical approach, but beware that using erosions as a surrogate marker for true ulcers is dangerous. This had led to many 'false positives' with respect to predicting the safety of NSAIDs when used chronically by patients (such as those with osteoarthritis). As Ierardi Enzo pointed out, there is emerging evidence that drugs that protect the upper GI tract from damage induced by NSAIDs just shift the damage to the more distal small intestine, where it is more difficult to detect and treat (thus, much more dangerous).

Endoscopy can not be described routinely to asses the NASIDS induced injuries. NASIDS induced injuries are not limited to the upper GIT, BUT GROWING EVIDENCE SUPPORT SMALL AND LARGE INTESTINAL INJURIES AS WELL. We may follow patients clinically by simply reporting new symptoms and postpone endoscopy for symptomatic cases.

There have been several papers with video capsule endoscopy (VCE) in NSAID users. This method appears to be promising for detecting the lesions in the more distal GI tract. However, it is expensive to do VCE, so unlikely that it will become a widespread approach for NSAID-enteropathy until the price becomes more reasonable. A non-invasive marker of small/large intestinal damage caused by NSAIDs would be very useful (if affordable), but so far none have been identified that are sufficiently sensitive, specific and simple to use.

The discussion is exciting and interesting. I have experienced NSAID intestinal damages detected by VCE and their frequency is impressive. However, I should like to invite all discussants to came back to the primary question of Thais. In patients who must assume NSAIDs for four weeks, how do you design a study with the aim of detecting gastric and intestinal damages? And in this not long period, how many patients will develop GI damages, i. e. has this kind of study a suitable rationale? I think that these questions have not been clearly answered despite the large partecipation to the discussion. I should like, finally, to know whether the study was performed or started or our opinions discouraged dr. Figueiredo to perform it.

The matter of time/ duration is a key point in this topic. I do not think that one can find any significant problem regarding the use of a 4 weeks period of NSAID use, by instinct. I am talking on evidence!

But if any erosion/ ulcer develops in a small subgroup of patients using NSAID, would it be clinically significant?

I would also like to know if the discussion is going on or the research has been conducted? Is so I am so curious on the results and I like to ask Thais to declare the situation of the discussion.

The study hasn´t started yet, it´s still being defined!

I think that a good approach could be an animal study, then you can use histology and Visual Analogue Scale to evaluate the damage caused by 2 different NSAIDs during a 4-week treatment.

The problem is not simple. First, I do not know whether there is an animal model able to mimic human GI lesions induced by NSAIDs. Second, where must histological evaluation be performed? Third, at the best of my knowledge, there are no specific NSAID induced histological alterations (non-specific inflammatory change may be due to a lot of causes).

The spectrum of damage caused by different NSAIDs range from microvillous injury to malabsorption, GI bleed and intestinal strictures. However in sort term use, GI bleed is most common side efect. I think appropriate symptoms, a fecal occult blood test followed by upper GI endoscopy if positive are the first line investigation for assessing damage