This means that in this class of obese people, if we operate on a thousand, two will die. If you take these same thousand people and just let them carry on the same way (with medical treatment, etc.) HALF of them would be dead in 20 years! In fact, the five year mortality of these patients is high enough to justify bariatric surgery.

Even though this is pretty impressive stuff to demonstrate how life-saving and life-altering bariatric surgery can be, what happens to an individual case where the patient dies? For that family, the mortality rate is 100%! Can things be any worse for them? No.

Therefore, if we could try and reduce the mortality of surgery even further, why would we not do it?

If we could weed out the highest risk cases, it would be half the battle won. We could then simply try not to operate on this high-risk class of patients, UNLESS there is a clear and explicit understanding and acceptance that they have a significant chance of dying within a month of surgery.

So the highest risk is seen in these groups:

1. Elderly
2. Males
3. High BMI
4. Smokers, alcoholics
5. History of pulmonary embolism/DVT
6. Re-do bariatric surgery

• Causes of 30-day bariatric surgery mortality: with emphasis on bypass obstruction– Mason EE, Renquist KE, Huang YH, Jamal M, Samuel I.
• The SOS Study: Sjöström
• The Pennsylvania bariatric mortality study
• Long-Term Mortality after Gastric Bypass Surgery: Ted D. Adams, et al.
• Medicolegal analysis of 100 malpractice claims against bariatric surgeons: Cottam D, Lord J, Dallal RM, Wolfe B, Higa K, McCauley K, Schauer P.

Click on pen to

This is an oversimplification, IMO.

I have mentioned before that the sleeve, in spite of being an easier and faster procedure, may be more problematic in terms of leak alone. The leak rate of the sleeve is around 2-3 percent, while that of the bypass is around the same, but the leaks behave differently.

The sleeve leak (as discussed in details before) is a high pressure leak, occurring usually at the angle of His, and tends to persist unless stented.

The bypass leak occurs at multiple possible sites, half in the gastrojejunostomy, but being a low pressure leak, tends to dry out on its own once the surgical re-exploration is done and issues sorted out.

(pic from referenced article)

Leaks after a bypass are usually detected on Day 3, give or take some. The detection may be because of routine dye study, but usually the patient already has a high pulse rate, fever, abdominal pain, dirty drain output, etc. It is important to remember that around a third of patients may have no clear evidence of a leak in spite of CT scan or dye test or both. In fact, in an older study, only a pulse rate over 120 and a rapid breathing rate were consistent in patients with leaks, and only 2 out of 9 leaks showed up on the dye study.  This means that a laparoscopic re-exploration is the most reliable of investigations for a leak!

Upon detection of a leak after the bypass, the patients is usually re-explored laparoscopically. The procedure may include any or all of the following:

1. Re-suture of leak with omental graft
2. Re-do the full anastomosis
3. Partial remnant gastrectomy (in case of remnant staple line leak)
4. Drainage
5. Gastrostomy

While around 2 percent of bypasses may leak, when they do occur, they inflict considerable collateral damage: gastrogastric fistula, wound complications, lung problems, DVT/PE, etc.

It is also important to remember that a downstream problem like an obstruction of the jejunum may cause leaks from the staple line above.

Are there groups which have a predictably higher leak rate?

Yes, it may be said that the leak risks are higher in:

1. Revisional bariatric surgeries (around 13-14%)
2. Males.
3. Higher BMI.
4. Multiple co-morbidities.
5. Patients with previous abdominal operations.
6. Circular stapled anastomoses (hand-sewn method has the lowest leak rate of 0.4%): controversial!
7. Inexperienced surgical hands.
8. Those who have had an intraoperative mishap (colon/splenic injury, for example).
9. One leak may predispose to another.
10. Buttressed staple lines may have an advantage in leak rates.

The mortality of jejuno-jejunal leaks is close to 40-50 percent, while a GJ leak has a 10% mortality.

What happens once a leak is detected and the patient re-explored? Initial days may be stormy or tense, spent in the ITU with ventilatory support given to some patients (remember, many of these patients are already sick with other co-morbidities), and then the sepsis gets controlled and the nutrition support started (through the gastrostomy tube or through a central vein in the neck, etc.). The patient recovers slowly and the leak heals with time. In the rare case, if a leak persists beyond a month, a stent may be required.

With this post, our continuing exploration of the Achilles heel of all bariatric operations has been, you will admit, rather thorough. If this and the other articles frighten you, then be assured that this was and is not the goal of my writing. We at BMI believe that an informed patient is the best one, and we owe it to you to even out the information asymmetry in this highly technical branch of medical care. All surgeries have side effects, but they occur in a small minority of patients. In the given case, each complication hits hard, and a prepared patient handles it better. As bariatric surgeons, we need to brief you adequately as to the possible downsides of what is otherwise a transformational event in life.

Ref:

Diagnosis and Contemporary Management of Anastomotic Leaks after Gastric Bypass for Obesity

J Am Coll Surg

Rodrigo Gonzalez, MD, Michael G Sarr, MD, FACS, C Daniel Smith, MD, FACS, Mercedeh Baghai, MD, Michael Kendrick, MD, Samuel Szomstein, MD, FACS, Raul Rosenthal, MD, FACS, Michel M Murr, MD, FACS

Click on pen to

I had jokingly noted how BMI has not had a leak in many years. This year (obviously our busiest year so far) we have had three leaks, each unique in mechanism, presentation and treatment.

I will discuss the cases separately later in my posts, and share some thoughts about this important topic.

Sleeve gastrectomy leaks behave totally differently compared to any other type. The reasons are multiple, and all lead to a persistent and nagging leak that may end up as a fistula (a tract between the stomach and the skin).

The causative factors are:

1. The pylorus (ring of muscle that acts a the gateway from the stomach into the duodenum) is normally closed, and the pressure in the gastric tube increases because of that, especially when a peristaltic wave builds up in the organ. The pressure may be as high as 45 mm Hg pressure, whereas the lower esophageal pressure is 20 mm Hg at most. The resultant gradient of pressure is from the pyloric end towards the esophagus. This leads to high pressure acting on the highest end of the staple line (near the GE junction). Greater the pressure, the higher the leak rate.
2. The GE junction area is least supplied by blood vessels. This would lead to relatively weaker healing.
3. The intrathoracic pressure is negative, and the pressure differential in the upper stomach, therefore, is high.
4. The gastric tube may twist in a corkscrew manner in the postoperative phase, perhaps as a result of the staple line running in different directions and the muscle layers getting transected at different levels.
5. The incisura of the stomach may be narrowed by the stapling. This would not only lead to high pressure build up in the stomach but also make leak management doubly difficult.

Whatever be the cause, leaks may increase mortality (10%), and certainly do increase hospitalisation and costs of treatment. In fact, in the Indian context, where insurance is nowhere in the picture, it is scary to imagine a leak in a patient with limited resources.

Some other points about sleeve leaks:

* The most leaks occur when both staple line buttresses and overrunning sutures are used (Gagner).

* Intersections of staple lines must always be oversewn to prevent leaks.

* The narrow sleeves leak the most compared to the wider ones (typically 32 Fr versus 40 Fr).

* Leaks may occur even as late as 2 to 4 weeks after operation, though this is not common.

* A leak mandates immediate re-exploration, drainage and insertion of a feeding tube (nasojejunal or jejunostomy or parenteral).  Many experts now routinely insert a stent intraoperatively during the re-exploration. This possibly reduces hospital stay.

* A chronic leak persisting over three months is not amenable to endoscopic stenting and would need surgery.

Endoscopic stent management of sleeve leaks (some major points to bear in mind):

For stents to work, the defect should be less than 3 cm or less than 50% of the circumference of the organ.

As mentioned above, the consensus seems to be emerging that leaks should be stented during the re-exploration. At this re-exploration, an attempt may be made to identify the leak, suture it with an omental pedicle as buttress and drain the area. If the leak is not easily visible, one should not hunt it, for fear of causing injuries in the friable and inflamed area.

If a stricture is present along with a leak, two stents may be used to cover the entire sleeve, or the single stent has to reach up to the incisura/stricture.

SEMS for Sleeve Leak (video)

The stents used may be Self Expanding Plastic Stents (SEPS) known as Polyflex or SEMS (self expanding metal stents). The problem with the Polyflex or plastic stents is that they tend to migrate, while the SEMS stents are very difficult to remove. To this end, the two stents have different problems and different solutions have been tried successfully: to prevent migration of the SEPS stent, a thread is passed through the edge of the stent and brought out through the nose and tied to the patient’s ears (!). Ingenious and cheap. With the SEMS stent, double coating may make for easier removal. I have seen one presenter at IFSO 2011 at Hamburg mention and show esophageal avulsion injury during removal of a metal stent.

Stents cause problems:

• Reflux
• Salivation in the morning
• Chest pain
• Dislodgment
• Erosion
• Difficult removal

Stenting should be accompanied by endoscopic pyloric dilatation, using 20 psi pressure. This reduces the intra-sleeve pressure.

Checking the stent position every week or month (depending on the policy) is necessary.

Endoscopic methods other than stents:

• For leaks less than 1 cm, several injections of fibrin sealant may lead to 100% success rates.
• For the same kind of leak, argon laser coagulation has also been used (Basso).
• Through-the-scope clips may be used for defects less than 1 cm.
• Over-the-scope clips may be used for larger defects.
• A Brazilian center advocates an endoscopic septotomy (division of a ridge like elevation that forms beside a chronic fistula) and cleaning up of the abscess adjoining the fistula, thereby leading to an internal drainage of the leak. This allows immediate removal of the drain. This is an unorthodox method of treatment.

Surgery:

For a chronic leak, surgery may be a last resort.

1. The simplest is to connect a jejunal loop to the fistula defect
2. The sleeve is converted to a gastric bypass and the fistula is left untouched. This converts the high pressure sleeve leak to a low pressure bypass leak. The fistula then dries out.
3. Conversion to bypass and stoma constructed at the leak site.
4. Total gastrectomy with esophagojejunostomy (Neto).

I hope this series has been educational for you, whether you are a patient, relative, doctor or student.

Click on pen to

Before we proceed, do you know what is a lap gastric bypass all about?

An important factor to be considered by every patient/doctor involved in a gastric bypass is how the stoma for the pouch would be created, as it is one of the known factors leading to weight regain.

The stoma is the opening we create between the gastric micro-pouch and the upper small bowel (jejunum). Too big a stoma will lead to food exiting the pouch quicker and resultant weight regain. Too small a stoma would lead to obstructive symptoms and misery.

So what is the ideal stoma size: 12 mm, 14 mm, 18 mm, or 25 mm? And how would we create this stoma? Does the stoma size stay that way all life at the same size?

Easy, tiger, easy. One thing at a time!

Ideal stoma size:
Different people seem to be using different sizes. A popular way of creating the stoma is with the EEA circular stapler and the OrVil device. The stoma created is either 21mm or 25mm in size.
The stoma is a little big but it is known to contract in size later. The risk of stenosis being highest in the circular stapler group, it is perhaps better to keep stoma size around this level. A smaller stoma may stenose and cause obstructive symptoms. The incidence of stenosis is less than 3 percent.

The hand-sutured technique (my personal favorite) is known to stenose the least in the long term. Here the stoma size is 12 mm/36 Fr. In case you are wondering how we can measure in millimeters in lap surgery, we pass a 36 Fr gastric tube through the mouth into the pouch. The anastomosis between the pouch and the jejunum is fashioned over this tube.
A stoma can be created with the Linear Stapler. This leaves the anterior (front) walls open, and need hand-suturing to complete the process. The stoma also tends to be excessively wide, in my experience. Therefore, at BMI, we don’t favor this method, preferring the hand-sewn and EEA techniques instead.
So the keen patient here would be wondering, “Which is better or best?”
Like most things in life, it depends. Each method of stoma creation has its pros and cons.
The EEA circular stapled method is quick and mechanised, so suturing is not a major issue. The stoma is reliably and safely created, with minimal leak rates. However, it is expensive, needs the surgeon to be familiar with the device, and has the highest stenosis rates. On top of that, technical disasters are not unknown (study in colorectal context).

The hand-sewn method (best demonstrated by Kelvin Higa) is also safe, reliable and avoids mechanical stapler failures and disasters. It needs the highest level of skills in the surgeon, and is more time consuming in most hands. Experts, however, can complete this quickly.

The Linear Stapler method is easy for most surgeons, but has very little to recommend it.

One final issue with the stoma: do we need stoma to be buttressed, or glued? It may by nice to do and for us to see in the operating room, but it probably does not reduce leak rates significantly. In the high-risk patient we may choose to use it. Some studies do report significant benefits in reduction of bleeding, leaks and stenoses.

The bottomline is to ask your surgeon how he would create the stoma, and what he prefers. Then make up your mind to go for it. You have more to gain by that decision: that one thing that will transform your life.

Click on pen to
]]> http://www.bmi-india.com/gastric-bypass-choosing-the-stoma/feed/ 0 http://www.bmi-india.com/does-bougie-size-in-sleeve-gastrectomy-matter/ http://www.bmi-india.com/does-bougie-size-in-sleeve-gastrectomy-matter/#comments Mon, 07 Feb 2011 22:57:50 +0000 ramana http://www.bmi-india.com/?p=377 When we started out on doing the sleeve gastrectomy a few years back, it was typically over a 60Fr bougie. The important thing, we learnt, was not to leave any part of the fundus of the stomach behind. This was the important technical thing to check. Our old patients have all done well.

Since the last couple of years, however, it has become commonplace to do tight sleeves over bougies 32-36 Fr. The idea is that a tighter sleeve is more restrictive in terms of stomach capacity and would give faster and better weight loss.

This article examines the truth.

Gagner et al had published an article in Obesity Surgery in 2008 showing better safety profile in their own series for patients who had 60 Fr sleeves compared to those who had narrower 40 Fr sleeves. Today, however, we are talking of much tighter sleeves. The question, therefore, could be “Does a sleeve twice as tight as a conventional 60 Fr sleeve give better results?”

Recent studies have shown that tight sleeves have shown good results. In fact, many of us in India have no regrets doing tight sleeves, with equivalent weight loss in the short term and 85-100% resolution of diabetes mellitus.

At BMI, we use around 34 Fr or 38 Fr, but end up with a slightly tighter sleeve because of meticulous oversewing of the staple line.

The most important consideration in designing a tight sleeve is to ensure that there is no narrowing at the level of the incisura (the sag-point of the stomach where the organ turns horizontally towards the right of the patient). A narrowing will cause a leak at the GE junction or lead to persistent vomiting later, necessitating invasive treatment later. All in all, a highly regrettable outcome could ensue, so it is wise to leave a wider incisura than make it too tight.

Some new thoughts are also currently being aired on improving outcomes after the sleeve. Could the gastric tube, detached from its greater curve attachment of gastrocolic (“greater”) omentum, get coiled up later and impair emptying of the organ? Would it be wise to re-anchor the omentum to the stomach tube?

We really don’t know yet. Interesting possibilities abound, but the truth is by no means established.

In recent times, several studies have reported no increase in complications  when the staple line is not reinforced with stitches. In fact the last one I read talked of zero leaks in  more than 500 cases. At BMI, we have now switched from ”meticulous suturing of the staple-line” above to stitches placed at the junction of adjacent staple lines.

And, so far, we are free from the disaster of leaks (knocks on wood).

So, if we create tighter sleeves with a narrow bougie (between 32 and 36 Fr) would we have a higher rate of strictures? No. In a recent article in a journal, Zundel et al have reported a 0.26 % stricture rate. The cause of stricture may be:

1. A hematoma (blood clot) causing compression

2. Stitch causing narrowing

3. Ischemia due to inadvertent vascular injury to branches of the left gastric artery.

4. Kinking of the gastric sleeve (due to stitches, stapling technical errors, etc). This can occur regardless of sleeve size.

5. Fistula: the associated inflammation causes narrowing.

6. Band removal cases, usually associated with much fibrosis.

7. Mucosal edema and food impaction can also cause acute blockage.

References:

1. Gagner M, Deitel M, Kalberer TL, et al. The Second International Consensus Summit for Sleeve Gastrectomy, March 19-21, 2009. Surg Obes Relat Dis. 2009;5:476–485.

2. Frezza EE, Reddy S, Gee LL, et al. Complications after sleeve gastrectomy for morbid obesity. Obes Surg. 19:684–687.

3. Lalor PF, Tucker ON, Szomstein S, et al. Complications after laparoscopic sleeve gastrectomy. Surg Obes Relat Dis. 2008;4: 33–38.

4. Zundel, et al. Strictures After Sleeve Gastrectomy.

Surg Laparosc Endosc Percutan Tec, Volume 20, Number 3, June 2010

5. DE Bellanger, FL Greenway. Laparoscopic Sleeve Gastrectomy: 529 cases without a leak.

Click on pen to
]]> http://www.bmi-india.com/does-bougie-size-in-sleeve-gastrectomy-matter/feed/ 3 http://www.bmi-india.com/gastric-plication-a-new-wls-procedure/ http://www.bmi-india.com/gastric-plication-a-new-wls-procedure/#comments Sun, 18 Jul 2010 04:22:12 +0000 ramana http://www.bmi-india.com/?p=456 Many bariatric surgeons are excited about the procedure Gastric Plication (LGP) as a bariatric procedure. It promises to be a simple method (remember, “Less Is More!”) that gives weight loss results as good as sleeve gastrectomy without even the risks of that procedure (staple line leaks (read part one and part two here) or bleeding).

What is done is that the greater curvature of the stomach is freed from the tissues attached to it (known as the gastrocolic omentum) using a vessel sealing device like the Harmonic Scalpel or the Ligasure. This is the first step of the sleeve gastrectomy procedure, as well.

However, unlike the sleeve, here we do not use the stapler to remove 80% of the stomach. We merely imbricate/plicate the stomach using running stitches from above downwards, layer by layer, till most of the stomach is pushed inwards, creating a narrowing of the passage, and a tunnel along the lesser curvature, just like the sleeve.


Lap Gastric Plication

In the same vein, if someone wants to reverse the procedure, the stitches may be removed by laparoscopy (making this a potentially reversible procedure like the Lap Band), though this would neither be a good thing nor a great experience!

Some patients have significant reflux symptoms after LGP, as after the sleeve, as we have discussed before.

It is possible that the stitches may give way later and hamper the weight loss, but preliminary results, as presented in the recent International Conference of Obesity in Stockholm, show good results akin to the sleeve.

More info later!

Click on pen to
]]> http://www.bmi-india.com/gastric-plication-a-new-wls-procedure/feed/ 45 http://www.bmi-india.com/leaks-after-sleeve-gastrectomy-part-two/ http://www.bmi-india.com/leaks-after-sleeve-gastrectomy-part-two/#comments Mon, 24 May 2010 14:41:02 +0000 ramana http://www.bmi-india.com/?p=379 In Part One, we came to the reluctant conclusion that the sleeve, safe procedure though it is, has a near 3 percent leak rate. Incidentally, the two of us at BMI have not yet had a leak in nearly five years of practice. For whatever reason though it may be (luck?), we would like to think it is because of our superior technique!



We need to now examine why these leaks occur. Is it the make of the stapler?

All over the world, surgeons use either the Ethicon stapler (called Echelon) or the Covidien product (the legendary endo-GIA). In our experience, while both are world class products, the mechanical problems arise with reuse of the stapler guns. In India, not many patients want to pay for a new gun. Surgeons also tend to give lower packages in an effort to be patient-friendly, thereby giving short shrift to a new gun (something that may cost nearly 20,000 to 30,000 INR, or in the vicinity of 400-500 USD).

If you reuse a gun beyond its shelf life, you can have misfirings during the operation. This is a ghastly experience which no surgeon should have. I have even had occasions where the stapler failed to unlock after firing! Clearly, no surgeon should allow a situation like this to happen, but how can he prevent it? After all, the stapler gun doesn’t say, “Don’t use me now, I am gonna die!”

What I say now may be controversial, but this is my humble experience of using laparoscopic staplers over a decade. The Covidien gun tends to work better with reuse, while the Echelon stapler works best when new. If I use the latter product in a case, I make sure I buy a new gun and throw it off (after all, these products were not meant to be reused!) after the case. Obviously, this pushes up the cost of the operation, but what can we do except to hope that the patient understands the reasons?

To come back to the point, reusing staplers is a tricky and potential troublesome issue. I have found no evidence in the scientific literature linking leak rates with reuse, but I suspect that there may be a relationship in some cases, at least.

In a personal communication with the authors of the Chilean paper I referenced in Part One, they opine that leak rates may be related to thermal injuries. This means that when we seal off and divide the blood vessels of the stomach, we may cause some heat injury to the wall of the organ. If such a part is left behind (rather than removed as part of the specimen) it may leak in the post-op period.

Another mechanism of leak: if the gastric tube is too narrow at the region of the body ( a point called the incisura) the resultant increase in pressure in the upper part of the stomach tube may lead to a blowout at the most vulnerable part above. This is usually at the junction of the food pipe and the stomach (the GE junction). This is borne out in clinical practice. The commonest site of a leak is the GE junction.



Another site of a leak could be the junction between adjacent staple cartridges. This is why it is considered important to oversew these junctions.

Does oversewing the staple line prevent leaks? We all think it  does, which is why practically all of us do so. However, as this Czech paper says, it may be unnecessary in most cases.

How do we detect leaks?

During the operation, we check the staple line by pushing in methylene blue dye into the stomach. A leak will be seen if present. Some people use an endoscopic verification of the staple line.

After the operation, a contrast (dye) study usually done just before liquid diet is started may be done, especially if intraoperative checks were not done, or a leak was detected and corrected at that time.




Does using staple line reinforcement reduce leaks? As this literature review says, no.



Which patients are more prone to leaks?

While the heaviest middle-aged male smokers are the stereotypical ‘bad’ patients, those who undergo revision surgery (for example, a sleeve with a band removal) are more liable to leaks.

How is a leak managed?

1. Ensure drainage of the peritoneal cavity (percutaneous, CT-guided drainage versus laparoscopic surgical placement of a drain)

2. Endoscopic stent to cover the leak, though a stent may migrate and be unsatisfactory in a given situation.

3. Suture closure of the leak after re-exploration of the abdomen.

4. Parenteral or enteral nutrition.

5. Mere observation in given patients

In most instances, the leak takes several weeks to dry out fully, and this results in prolonged hospitalisation and increased costs as well.

The important thing about leaks is to detect it early, as clinical examination in the severely obese is notoriously unreliable.




Useful references:

1. ASMBS position paper on Sleeve Gastrectomy 2009

2. Canadian overview on Sleeve Gastrectomy

3. World Journal of Gastroenterology 2008 Editorial

Click on pen to
]]> http://www.bmi-india.com/leaks-after-sleeve-gastrectomy-part-two/feed/ 4 http://www.bmi-india.com/leaks-after-sleeve-gastrectomy-part-one/ http://www.bmi-india.com/leaks-after-sleeve-gastrectomy-part-one/#comments Mon, 17 May 2010 03:58:01 +0000 ramana http://www.bmi-india.com/?p=370 Our patients are counseled in details about the pros and cons of various bariatric procedures, especially the one they are going in for. In today’s practice, most of our patients tend to favor the Sleeve Gastrectomy for its safety, ease of maintenance and less restrictive lifestyle. It also may be cheaper than the bypass to variable extent.

We as surgeons tend to counsel patients according to our own perspectives. I have never failed to acknowledge that, while I can place in a Band as well as another Johnnie, I have a distinct distaste for it. My counseling tends to betray this bias. I am objective enough to acknowledge this, while many others would put an evidence-based spin to their own colored viewpoint.

So our pre-op counseling tends to favor the sleeve. I do try to attract the patient to the benefits of the bypass, but I am careful when doing so. An inappropriate procedure in an unsuited patient can be a miserable experience. As I was saying before interrupting myself for the nth time, we tend to portray the sleeve as the safest procedure for the patient, with very negligible leaks and problems like bleeding. We tend to convey the impression that the expected complications are more likely to be those of any procedure in the severely obese patient, like embolism, pneumonia, infections, etc.

Now, if I can be brutally frank about this, we are not being entirely factual. Why? Because even an operation as safe as sleeve gastrectomy does have a specific leak rate in the literature. How much? If you look at a recent prospective study from the famous Chilean University Hospital known for the great surgeon Atilla Csendes, they had seven leaks in 214 patients, around 3 percent. This is higher than other papers like this one from Cleveland Clinic, where the leak rate was 0.7 percent. However, on the whole, a figure of 2.7 percent is an accepted leak rate arrived at from 24 studies covering over 1700 patients. To our surprise, we find that the leak rate of the sleeve may be more than after the bypass, an operation generally acknowledged as having more complication rates (Nguyen et al).

There are several major issues of import here:

Are the leaks stapler related?

Are these leaks reduced by staple line reinforcement methods like suturing or Seamguard?

Can we identify a subset of patients who are more susceptible to leaks?

How do we manage these leaks?

For all these and more, stay tuned for Part Two.

Click on pen to
]]> http://www.bmi-india.com/leaks-after-sleeve-gastrectomy-part-one/feed/ 4