Data-mining studies have limitations
Article Type
Changed
Wed, 01/02/2019 - 09:42


Laparoscopic sleeve gastrectomy is safe and effective overall, but staple line reinforcement appears to increase the rate of postsurgical leaks – which were associated with readmissions and, in some cases, reoperations.

A large review of quality improvement data found that staple line reinforcement – an extremely common technique – was associated with a 60% increased risk of leak, compared with closures without staple line reinforcement, Elizabeth R. Berger, MD, and her colleagues reported in the October issue of the Annals of Surgery (2016;264:464-73).

“This study also demonstrates that leaks were significantly more morbid than bleeding with higher readmission and reoperation rates in patients with a leak vs. a bleed,” wrote Dr. Berger of Loyola University, Chicago, and her coauthors. “Therefore, a surgeon should consider the benefits, risks, and costs of each surgical technique in performing a laparoscopic sleeve gastrectomy and selectively utilize those that, in their hands, minimize morbidity while maximizing clinical effectiveness.”

The team examined outcomes in 189,477 laparoscopic sleeve gastrectomies performed by 1,634 surgeons at 720 centers from 2012 to 2014. All of the data were extracted from the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program, created in 2012 by the American College of Surgeons and the American Society for Metabolic and Bariatric Surgery.

They examined the impact of staple line reinforcement, oversewing the staple line, bougie size, and distance of the staple line from the pylorus on 30-day outcomes, and their effect on weight loss and weight-related comorbidities at 1 year. Outcomes included morbidity, leak rates, and bleeding, which were examined at both the patient and surgeon levels.

Most patients (126,904; 67%) patients had some type of staple line reinforcement (SLR); the rest had only oversewn staple lines (OSL) or no reinforcement. Leaks occurred in 1,703 patients and bleeds in 1,436 patients. Leaks were more serious than bleeds: Patients with a leak were almost 28% more likely to readmitted and 11% more likely to need a reoperation than were patients who had only a bleed.

At the patient level, those with SLR with or without OSL were 20%-46% more likely to experience a leak than were those who had neither. Bleeding rates were about 70% lower in the SLR groups.

Most surgeons in the analysis (80%) used some type of SLR; almost 20% routinely used only OSL, and 30% routinely used only SLR. At the surgeon level, SLR was associated with a 60% increased risk of a postoperative leak, compared with no reinforcement. There was no association between SLR and bleeding risk, however.

Oversewing had an effect on 1-year weight loss. Patients with oversewn staple lines lost an additional 1.3 points on the body mass index (BMI) scale, compared with patients with no type of reinforcement.

“The reason for increased leaks from SLR is relatively unclear,” the authors wrote. “The two layers of material that are placed within the staple line could increase ischemia or decrease the relative staple heights. At the notches, where one staple firing ends and the next one begins, there is sandwiching of the two layers of staples and a combined four layers of SLR. This bulk may predispose to leaks.”

Larger bougie sizes (BS) seemed more beneficial than did smaller ones, in both the surgeon- and patient-level analyses. A BS of at least 38 French was associated with a 28% decreased risk of a leak (odds ratio 0.72) at the patient level and a 10% decreased risk at the surgeon level (OR 0.90). There were no associations with bleeding.

“Our findings support literature that describes narrower BSs leads to increased ischemia secondary to increased intraluminal pressure, causing more leaks,” the authors wrote.

A BS of at least 40 French had a significant impact on weight loss. At 1 year, patients with the larger BS had lost 2.45 points more on the BMI scale than did those with smaller sizes.

This finding is in accord with other studies, including one that found the best weight-loss outcomes associated with a BS of more than 60 French. “Perhaps the sleeve works because of more rapid emptying, which is favored by a relatively larger BS, rather than because of restriction,” they said.

The distance to the pylorus (DP) from the staple line initiation point was divided into four sections: less than 4 cm; 4-5 cm; 5-6 cm; and 6 cm or more.

On a patient level, there was no association between DP and leak rates. There was, however, an association with bleeding. A DP of 4-4.99 cm had the highest rate, 90%, while a DP of 5-5.99 cm had the lowest (71%). DP was also associated with weight loss on this level, with a distance of more than 6 cm being associated with the biggest BMI decrease (3.7 points).

“Our data show significantly increased excess weight loss in a stepwise fashion as the DP increases,” the authors said. “Our data suggest that as DP increased, there was an increased excess weight loss, possibly explained by preserving the ‘antral mill.’ Stapling further from the pylorus perhaps keeps the antrum’s functional component intact and allows food to enter the distal gut more quickly, leading to earlier satiety and increased weight loss.”

Only 114 surgeons (8%) used a DP of less than 4 cm. There were no significant associations with any 30-day outcomes and DP after adjustment.

The authors had no financial disclosures.

 

 

Body

Before drawing overarching conclusions and implementing recommendations based on this study, there are several limitations that must be borne in mind when considering data-mining exercises such as this one:

• It should be taken into account that there was significant intraoperative variation in technique and experience among the surgeons that was not captured through the data acquisition.
• Similarly, the true distance between the stapler and the selected bougie is also variable, adding an inherent lack of accuracy of the true real diameter of the completed gastric tube.
• There is a lack of granular information, including the type of SLR or staplers used, thereby also limiting any reliable conclusions that could be drawn.
• There are additional techniques, such as omental buttressing, and the use of clips, sutures, or hemostatic agents that are not reported, yet may have an impact on leak and bleeding rates.
• The reported follow-up rate of 39.4% at 1 year is typically considered to be suboptimal.
• SLR techniques may also include oversewing, and these are also subject to wide variation, including the type of suture material used, and the actual suturing technique that was implemented.
• Only those patients whose bleeding was severe enough to warrant transfusions were included, such that lower level bleeding would have not been represented in this report.
• There were also deficiencies in correlating leaks or bleeding rates with staple height selection, or the experience and learning curve of the surgeon.

Dr. Samer Mattar
It is therefore incumbent on the reader to accept data-mining reports such as this one with a critical eye, taking into consideration that such papers present the readers with observations, but it would be inadvisable, if not hazardous, to draw conclusions and implement changes to current practice. Such actions are best guided by results from prospective, well-designed, comparative studies.

Samer Mattar, MD, is a bariatric surgeon and professor of surgery at Oregon Health and Science University, Portland. Dr. Mattar has no disclosures.

Publications
Topics
Sections
Body

Before drawing overarching conclusions and implementing recommendations based on this study, there are several limitations that must be borne in mind when considering data-mining exercises such as this one:

• It should be taken into account that there was significant intraoperative variation in technique and experience among the surgeons that was not captured through the data acquisition.
• Similarly, the true distance between the stapler and the selected bougie is also variable, adding an inherent lack of accuracy of the true real diameter of the completed gastric tube.
• There is a lack of granular information, including the type of SLR or staplers used, thereby also limiting any reliable conclusions that could be drawn.
• There are additional techniques, such as omental buttressing, and the use of clips, sutures, or hemostatic agents that are not reported, yet may have an impact on leak and bleeding rates.
• The reported follow-up rate of 39.4% at 1 year is typically considered to be suboptimal.
• SLR techniques may also include oversewing, and these are also subject to wide variation, including the type of suture material used, and the actual suturing technique that was implemented.
• Only those patients whose bleeding was severe enough to warrant transfusions were included, such that lower level bleeding would have not been represented in this report.
• There were also deficiencies in correlating leaks or bleeding rates with staple height selection, or the experience and learning curve of the surgeon.

Dr. Samer Mattar
It is therefore incumbent on the reader to accept data-mining reports such as this one with a critical eye, taking into consideration that such papers present the readers with observations, but it would be inadvisable, if not hazardous, to draw conclusions and implement changes to current practice. Such actions are best guided by results from prospective, well-designed, comparative studies.

Samer Mattar, MD, is a bariatric surgeon and professor of surgery at Oregon Health and Science University, Portland. Dr. Mattar has no disclosures.

Body

Before drawing overarching conclusions and implementing recommendations based on this study, there are several limitations that must be borne in mind when considering data-mining exercises such as this one:

• It should be taken into account that there was significant intraoperative variation in technique and experience among the surgeons that was not captured through the data acquisition.
• Similarly, the true distance between the stapler and the selected bougie is also variable, adding an inherent lack of accuracy of the true real diameter of the completed gastric tube.
• There is a lack of granular information, including the type of SLR or staplers used, thereby also limiting any reliable conclusions that could be drawn.
• There are additional techniques, such as omental buttressing, and the use of clips, sutures, or hemostatic agents that are not reported, yet may have an impact on leak and bleeding rates.
• The reported follow-up rate of 39.4% at 1 year is typically considered to be suboptimal.
• SLR techniques may also include oversewing, and these are also subject to wide variation, including the type of suture material used, and the actual suturing technique that was implemented.
• Only those patients whose bleeding was severe enough to warrant transfusions were included, such that lower level bleeding would have not been represented in this report.
• There were also deficiencies in correlating leaks or bleeding rates with staple height selection, or the experience and learning curve of the surgeon.

Dr. Samer Mattar
It is therefore incumbent on the reader to accept data-mining reports such as this one with a critical eye, taking into consideration that such papers present the readers with observations, but it would be inadvisable, if not hazardous, to draw conclusions and implement changes to current practice. Such actions are best guided by results from prospective, well-designed, comparative studies.

Samer Mattar, MD, is a bariatric surgeon and professor of surgery at Oregon Health and Science University, Portland. Dr. Mattar has no disclosures.

Title
Data-mining studies have limitations
Data-mining studies have limitations


Laparoscopic sleeve gastrectomy is safe and effective overall, but staple line reinforcement appears to increase the rate of postsurgical leaks – which were associated with readmissions and, in some cases, reoperations.

A large review of quality improvement data found that staple line reinforcement – an extremely common technique – was associated with a 60% increased risk of leak, compared with closures without staple line reinforcement, Elizabeth R. Berger, MD, and her colleagues reported in the October issue of the Annals of Surgery (2016;264:464-73).

“This study also demonstrates that leaks were significantly more morbid than bleeding with higher readmission and reoperation rates in patients with a leak vs. a bleed,” wrote Dr. Berger of Loyola University, Chicago, and her coauthors. “Therefore, a surgeon should consider the benefits, risks, and costs of each surgical technique in performing a laparoscopic sleeve gastrectomy and selectively utilize those that, in their hands, minimize morbidity while maximizing clinical effectiveness.”

The team examined outcomes in 189,477 laparoscopic sleeve gastrectomies performed by 1,634 surgeons at 720 centers from 2012 to 2014. All of the data were extracted from the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program, created in 2012 by the American College of Surgeons and the American Society for Metabolic and Bariatric Surgery.

They examined the impact of staple line reinforcement, oversewing the staple line, bougie size, and distance of the staple line from the pylorus on 30-day outcomes, and their effect on weight loss and weight-related comorbidities at 1 year. Outcomes included morbidity, leak rates, and bleeding, which were examined at both the patient and surgeon levels.

Most patients (126,904; 67%) patients had some type of staple line reinforcement (SLR); the rest had only oversewn staple lines (OSL) or no reinforcement. Leaks occurred in 1,703 patients and bleeds in 1,436 patients. Leaks were more serious than bleeds: Patients with a leak were almost 28% more likely to readmitted and 11% more likely to need a reoperation than were patients who had only a bleed.

At the patient level, those with SLR with or without OSL were 20%-46% more likely to experience a leak than were those who had neither. Bleeding rates were about 70% lower in the SLR groups.

Most surgeons in the analysis (80%) used some type of SLR; almost 20% routinely used only OSL, and 30% routinely used only SLR. At the surgeon level, SLR was associated with a 60% increased risk of a postoperative leak, compared with no reinforcement. There was no association between SLR and bleeding risk, however.

Oversewing had an effect on 1-year weight loss. Patients with oversewn staple lines lost an additional 1.3 points on the body mass index (BMI) scale, compared with patients with no type of reinforcement.

“The reason for increased leaks from SLR is relatively unclear,” the authors wrote. “The two layers of material that are placed within the staple line could increase ischemia or decrease the relative staple heights. At the notches, where one staple firing ends and the next one begins, there is sandwiching of the two layers of staples and a combined four layers of SLR. This bulk may predispose to leaks.”

Larger bougie sizes (BS) seemed more beneficial than did smaller ones, in both the surgeon- and patient-level analyses. A BS of at least 38 French was associated with a 28% decreased risk of a leak (odds ratio 0.72) at the patient level and a 10% decreased risk at the surgeon level (OR 0.90). There were no associations with bleeding.

“Our findings support literature that describes narrower BSs leads to increased ischemia secondary to increased intraluminal pressure, causing more leaks,” the authors wrote.

A BS of at least 40 French had a significant impact on weight loss. At 1 year, patients with the larger BS had lost 2.45 points more on the BMI scale than did those with smaller sizes.

This finding is in accord with other studies, including one that found the best weight-loss outcomes associated with a BS of more than 60 French. “Perhaps the sleeve works because of more rapid emptying, which is favored by a relatively larger BS, rather than because of restriction,” they said.

The distance to the pylorus (DP) from the staple line initiation point was divided into four sections: less than 4 cm; 4-5 cm; 5-6 cm; and 6 cm or more.

On a patient level, there was no association between DP and leak rates. There was, however, an association with bleeding. A DP of 4-4.99 cm had the highest rate, 90%, while a DP of 5-5.99 cm had the lowest (71%). DP was also associated with weight loss on this level, with a distance of more than 6 cm being associated with the biggest BMI decrease (3.7 points).

“Our data show significantly increased excess weight loss in a stepwise fashion as the DP increases,” the authors said. “Our data suggest that as DP increased, there was an increased excess weight loss, possibly explained by preserving the ‘antral mill.’ Stapling further from the pylorus perhaps keeps the antrum’s functional component intact and allows food to enter the distal gut more quickly, leading to earlier satiety and increased weight loss.”

Only 114 surgeons (8%) used a DP of less than 4 cm. There were no significant associations with any 30-day outcomes and DP after adjustment.

The authors had no financial disclosures.

 

 


Laparoscopic sleeve gastrectomy is safe and effective overall, but staple line reinforcement appears to increase the rate of postsurgical leaks – which were associated with readmissions and, in some cases, reoperations.

A large review of quality improvement data found that staple line reinforcement – an extremely common technique – was associated with a 60% increased risk of leak, compared with closures without staple line reinforcement, Elizabeth R. Berger, MD, and her colleagues reported in the October issue of the Annals of Surgery (2016;264:464-73).

“This study also demonstrates that leaks were significantly more morbid than bleeding with higher readmission and reoperation rates in patients with a leak vs. a bleed,” wrote Dr. Berger of Loyola University, Chicago, and her coauthors. “Therefore, a surgeon should consider the benefits, risks, and costs of each surgical technique in performing a laparoscopic sleeve gastrectomy and selectively utilize those that, in their hands, minimize morbidity while maximizing clinical effectiveness.”

The team examined outcomes in 189,477 laparoscopic sleeve gastrectomies performed by 1,634 surgeons at 720 centers from 2012 to 2014. All of the data were extracted from the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program, created in 2012 by the American College of Surgeons and the American Society for Metabolic and Bariatric Surgery.

They examined the impact of staple line reinforcement, oversewing the staple line, bougie size, and distance of the staple line from the pylorus on 30-day outcomes, and their effect on weight loss and weight-related comorbidities at 1 year. Outcomes included morbidity, leak rates, and bleeding, which were examined at both the patient and surgeon levels.

Most patients (126,904; 67%) patients had some type of staple line reinforcement (SLR); the rest had only oversewn staple lines (OSL) or no reinforcement. Leaks occurred in 1,703 patients and bleeds in 1,436 patients. Leaks were more serious than bleeds: Patients with a leak were almost 28% more likely to readmitted and 11% more likely to need a reoperation than were patients who had only a bleed.

At the patient level, those with SLR with or without OSL were 20%-46% more likely to experience a leak than were those who had neither. Bleeding rates were about 70% lower in the SLR groups.

Most surgeons in the analysis (80%) used some type of SLR; almost 20% routinely used only OSL, and 30% routinely used only SLR. At the surgeon level, SLR was associated with a 60% increased risk of a postoperative leak, compared with no reinforcement. There was no association between SLR and bleeding risk, however.

Oversewing had an effect on 1-year weight loss. Patients with oversewn staple lines lost an additional 1.3 points on the body mass index (BMI) scale, compared with patients with no type of reinforcement.

“The reason for increased leaks from SLR is relatively unclear,” the authors wrote. “The two layers of material that are placed within the staple line could increase ischemia or decrease the relative staple heights. At the notches, where one staple firing ends and the next one begins, there is sandwiching of the two layers of staples and a combined four layers of SLR. This bulk may predispose to leaks.”

Larger bougie sizes (BS) seemed more beneficial than did smaller ones, in both the surgeon- and patient-level analyses. A BS of at least 38 French was associated with a 28% decreased risk of a leak (odds ratio 0.72) at the patient level and a 10% decreased risk at the surgeon level (OR 0.90). There were no associations with bleeding.

“Our findings support literature that describes narrower BSs leads to increased ischemia secondary to increased intraluminal pressure, causing more leaks,” the authors wrote.

A BS of at least 40 French had a significant impact on weight loss. At 1 year, patients with the larger BS had lost 2.45 points more on the BMI scale than did those with smaller sizes.

This finding is in accord with other studies, including one that found the best weight-loss outcomes associated with a BS of more than 60 French. “Perhaps the sleeve works because of more rapid emptying, which is favored by a relatively larger BS, rather than because of restriction,” they said.

The distance to the pylorus (DP) from the staple line initiation point was divided into four sections: less than 4 cm; 4-5 cm; 5-6 cm; and 6 cm or more.

On a patient level, there was no association between DP and leak rates. There was, however, an association with bleeding. A DP of 4-4.99 cm had the highest rate, 90%, while a DP of 5-5.99 cm had the lowest (71%). DP was also associated with weight loss on this level, with a distance of more than 6 cm being associated with the biggest BMI decrease (3.7 points).

“Our data show significantly increased excess weight loss in a stepwise fashion as the DP increases,” the authors said. “Our data suggest that as DP increased, there was an increased excess weight loss, possibly explained by preserving the ‘antral mill.’ Stapling further from the pylorus perhaps keeps the antrum’s functional component intact and allows food to enter the distal gut more quickly, leading to earlier satiety and increased weight loss.”

Only 114 surgeons (8%) used a DP of less than 4 cm. There were no significant associations with any 30-day outcomes and DP after adjustment.

The authors had no financial disclosures.

 

 

Publications
Publications
Topics
Article Type
Click for Credit Status
Ready
Sections
Article Source

FROM THE ANNALS OF SURGERY

Disallow All Ads
Vitals

Key clinical point: Staple line reinforcement in laparoscopic sleeve gastrectomy was associated with increased leak rates.

Major finding: Compared to not reinforcing the staple line, doing sow as associated with up to a 60% increase in the risk of a postsurgical leak.

Data source: The database review contained outcomes on 189,477 laparoscopic sleeve gastrectomies.

Disclosures: None of the study authors had any financial disclosures.