Outcomes Data Used to Assess Residents' Surgical Skills

SAN FRANCISCO – Resident involvement in surgical procedures does not clinically affect surgical outcomes, according to a retrospective study of more than 60,000 cases from the National Surgical Quality Improvement Program database.

"There is a small – although questionable as clinically relevant – overall increase in mild and surgical complications. This is mostly caused by superficial wound infections when residents participate in surgical procedures," said Dr. P. Ravi Kiran, staff surgeon and head of the research section in the department of colorectal surgery at the Cleveland Clinic.

Using data from the National Surgical Quality Improvement Program database from 2005 to 2007, Dr. Kiran and his colleagues compared outcomes for patients who underwent surgery with and without resident participation.

The database, which includes data from pre-, intra-, and postoperative phases, uses clearly defined parameters and specialist nurse reviewers. It also includes resident participation and a morbidity probability, which offers an opportunity to use preoperative factors to stratify risk within subgroups, Dr. Kiran said at the annual meeting of the American Surgical Association.

Resident cases were matched with nonresident cases on the basis of age, sex, specialty, surgical procedure, morbidity probability, and important comorbidities and risk factors. Primary outcomes included 30-day mortality and postoperative complications (mild vs. severe, and surgical vs. medical). Secondary outcomes included the duration of surgery and length of hospital stay.

Mild complications included superficial surgical site infections (SSIs), peripheral nerve injury, urinary tract infection, deep venous thrombosis, and thrombophlebitis. Severe complications included deep (organ) SSI, wound disruption, bleeding requiring transfusion, failure of graft or prosthesis, reoperation, pneumonia, pulmonary embolism, acute renal failure, stroke, myocardial infarction, and sepsis.

Surgical complications included superficial SSI, deep (organ) SSI, wound disruption, bleeding requiring transfusion, failure of graft or prosthesis, peripheral nerve injury, and reoperation. Medical complications included pneumonia, pulmonary embolism, acute renal failure, stroke, myocardial infarction, sepsis, urinary tract infection, deep vein thrombosis, and thrombophlebitis.

For cases with resident vs. nonresident participation, the surgical complication rates were 7% and 6.2%, respectively – a significant difference – and mild complications rates were 4.4% and 3.5%, respectively. In addition, the mean operative time was significantly greater for cases involving residents – 122 vs. 97 minutes. The length of postoperative hospital stay was not significantly longer in the resident group.

The researchers identified 40,474 patients in the resident group and 20,237 patients in the nonresident group. The two groups were similar in terms of median age (50 years), sex (67% female), mean morbidity probability (0.09), American Society of Anesthesiologists classification, and presence of diabetes (6.4%) and hypertension (35%).

The groups were also similar in terms of presence of chronic obstructive pulmonary disease (0.27%), congestive heart failure or myocardial infarction in the past 6 months (0%), dialysis (0.044%), and preoperative sepsis (0.035%). Surgeons’ speciality areas were likewise similar for the two groups (general, 93%; vascular, 6%; and other, 1.36%).

Postgraduate year (PGY) 1-2 residents participated in 31% of operations, PGY 3-5 residents participated in 56%, and residents in PGY 6 or higher participated in 13% of cases.

The 10 most common surgical procedures were laparoscopic appendectomy, laparoscopic gastric bypass, laparoscopic cholecystectomy with and without operative cholangiogram, open appendectomy (nonruptured), thromboendarterectomy, colectomy (partial with anastomosis), laparoscopic colectomy (partial with anastomosis), ventral hernia repair, and placement of gastric band. These procedures were similar in terms of the percentages of resident and nonresident participation.

"We found that there was no difference in the [overall] 30-day mortality between the groups – 0.18% in the resident group and 0.20% in the no-resident group," said Dr. Kiran. However, any 30-day complications were 7.5% in the resident group and 6.7% in the nonresident group, a significant difference.

"When we further looked at the surgical complications, we noted that the cause of the difference in surgical complications between the two groups was the higher rate of SSIs in the resident group, when compared with the no-resident group ... the other surgical complications were similar," he said. The SSI rate was 3.0% for the resident group, compared with 2.2% for the nonresident group.

Interestingly, the researchers also found that overall 30-day complication rates increased with PGY – the rates were 6% for PGY 1-2, 8% for PGY 3-5, and 9% for PGY of 6 or more.

When they examined specific outcomes and complications between different PGY groups and matched cases without the involvement of residents, they found a similar pattern for the overall cohort.

"The reason for the difference in 30-day complications in the groups was because of differences in complications that were classified as mild, and primarily because the superficial surgical site infections were higher in the PGY 1-2 years, with an increased operative time," they said. The same was true for PGY 3-5 and PGY 6 and greater.

Also, as PGY increased, so did operative time – in both resident and nonresident groups. "This suggests that the reason for the increasing complications with increasing PGY years may have been related to increasing complexity of surgery," said Dr. Kiran.

"One overarching issue seems to be how we might achieve high-quality patient care and delivery of the clinical outcomes in the context of training," said Dr. Clifford Ko, a discussant.

However, he also acknowledged that teaching residents takes time. Dr. Ko, a colorectal surgeon and the director of the Center for Surgical Outcomes and Quality at the University of California, Los Angeles, questioned whether the longer operating time associated with resident involvement should be reduced.

"Although we would not perhaps be able to minimize time differences, I think that we have already achieved some mark of control by the gradation of responsibility over time, as residents continue with their training," Dr. Kiran said.

Although the surgical and mild complication rates were slightly greater, it’s unclear whether these differences are clinically relevant.

"The reasons for [these differences] are likely multifactorial and may be related to prolonged operative time. Considering that more complex cases may be performed in teaching hospitals and require resident participation, ‘resident’ could be a surrogate of severity of disease and intensity of operation – factors that may not be clearly discernible in a retrospective study – and this may explain the differences seen.

"Also, quality measures currently underway to reduce surgical site infections across the board may further minimize any of these differences that may exist," Dr. Kiran concluded.

The authors reported that they had no relevant disclosures.

SAN FRANCISCO – Resident involvement in surgical procedures does not clinically affect surgical outcomes, according to a retrospective study of more than 60,000 cases from the National Surgical Quality Improvement Program database.

"There is a small – although questionable as clinically relevant – overall increase in mild and surgical complications. This is mostly caused by superficial wound infections when residents participate in surgical procedures," said Dr. P. Ravi Kiran, staff surgeon and head of the research section in the department of colorectal surgery at the Cleveland Clinic.

Using data from the National Surgical Quality Improvement Program database from 2005 to 2007, Dr. Kiran and his colleagues compared outcomes for patients who underwent surgery with and without resident participation.

The database, which includes data from pre-, intra-, and postoperative phases, uses clearly defined parameters and specialist nurse reviewers. It also includes resident participation and a morbidity probability, which offers an opportunity to use preoperative factors to stratify risk within subgroups, Dr. Kiran said at the annual meeting of the American Surgical Association.

Resident cases were matched with nonresident cases on the basis of age, sex, specialty, surgical procedure, morbidity probability, and important comorbidities and risk factors. Primary outcomes included 30-day mortality and postoperative complications (mild vs. severe, and surgical vs. medical). Secondary outcomes included the duration of surgery and length of hospital stay.

Mild complications included superficial surgical site infections (SSIs), peripheral nerve injury, urinary tract infection, deep venous thrombosis, and thrombophlebitis. Severe complications included deep (organ) SSI, wound disruption, bleeding requiring transfusion, failure of graft or prosthesis, reoperation, pneumonia, pulmonary embolism, acute renal failure, stroke, myocardial infarction, and sepsis.

Surgical complications included superficial SSI, deep (organ) SSI, wound disruption, bleeding requiring transfusion, failure of graft or prosthesis, peripheral nerve injury, and reoperation. Medical complications included pneumonia, pulmonary embolism, acute renal failure, stroke, myocardial infarction, sepsis, urinary tract infection, deep vein thrombosis, and thrombophlebitis.

For cases with resident vs. nonresident participation, the surgical complication rates were 7% and 6.2%, respectively – a significant difference – and mild complications rates were 4.4% and 3.5%, respectively. In addition, the mean operative time was significantly greater for cases involving residents – 122 vs. 97 minutes. The length of postoperative hospital stay was not significantly longer in the resident group.

The researchers identified 40,474 patients in the resident group and 20,237 patients in the nonresident group. The two groups were similar in terms of median age (50 years), sex (67% female), mean morbidity probability (0.09), American Society of Anesthesiologists classification, and presence of diabetes (6.4%) and hypertension (35%).

The groups were also similar in terms of presence of chronic obstructive pulmonary disease (0.27%), congestive heart failure or myocardial infarction in the past 6 months (0%), dialysis (0.044%), and preoperative sepsis (0.035%). Surgeons’ speciality areas were likewise similar for the two groups (general, 93%; vascular, 6%; and other, 1.36%).

Postgraduate year (PGY) 1-2 residents participated in 31% of operations, PGY 3-5 residents participated in 56%, and residents in PGY 6 or higher participated in 13% of cases.

The 10 most common surgical procedures were laparoscopic appendectomy, laparoscopic gastric bypass, laparoscopic cholecystectomy with and without operative cholangiogram, open appendectomy (nonruptured), thromboendarterectomy, colectomy (partial with anastomosis), laparoscopic colectomy (partial with anastomosis), ventral hernia repair, and placement of gastric band. These procedures were similar in terms of the percentages of resident and nonresident participation.

"We found that there was no difference in the [overall] 30-day mortality between the groups – 0.18% in the resident group and 0.20% in the no-resident group," said Dr. Kiran. However, any 30-day complications were 7.5% in the resident group and 6.7% in the nonresident group, a significant difference.

"When we further looked at the surgical complications, we noted that the cause of the difference in surgical complications between the two groups was the higher rate of SSIs in the resident group, when compared with the no-resident group ... the other surgical complications were similar," he said. The SSI rate was 3.0% for the resident group, compared with 2.2% for the nonresident group.

Interestingly, the researchers also found that overall 30-day complication rates increased with PGY – the rates were 6% for PGY 1-2, 8% for PGY 3-5, and 9% for PGY of 6 or more.

When they examined specific outcomes and complications between different PGY groups and matched cases without the involvement of residents, they found a similar pattern for the overall cohort.

"The reason for the difference in 30-day complications in the groups was because of differences in complications that were classified as mild, and primarily because the superficial surgical site infections were higher in the PGY 1-2 years, with an increased operative time," they said. The same was true for PGY 3-5 and PGY 6 and greater.

Also, as PGY increased, so did operative time – in both resident and nonresident groups. "This suggests that the reason for the increasing complications with increasing PGY years may have been related to increasing complexity of surgery," said Dr. Kiran.

"One overarching issue seems to be how we might achieve high-quality patient care and delivery of the clinical outcomes in the context of training," said Dr. Clifford Ko, a discussant.

However, he also acknowledged that teaching residents takes time. Dr. Ko, a colorectal surgeon and the director of the Center for Surgical Outcomes and Quality at the University of California, Los Angeles, questioned whether the longer operating time associated with resident involvement should be reduced.

"Although we would not perhaps be able to minimize time differences, I think that we have already achieved some mark of control by the gradation of responsibility over time, as residents continue with their training," Dr. Kiran said.

Although the surgical and mild complication rates were slightly greater, it’s unclear whether these differences are clinically relevant.

"The reasons for [these differences] are likely multifactorial and may be related to prolonged operative time. Considering that more complex cases may be performed in teaching hospitals and require resident participation, ‘resident’ could be a surrogate of severity of disease and intensity of operation – factors that may not be clearly discernible in a retrospective study – and this may explain the differences seen.

"Also, quality measures currently underway to reduce surgical site infections across the board may further minimize any of these differences that may exist," Dr. Kiran concluded.

The authors reported that they had no relevant disclosures.

SAN FRANCISCO – Resident involvement in surgical procedures does not clinically affect surgical outcomes, according to a retrospective study of more than 60,000 cases from the National Surgical Quality Improvement Program database.

"There is a small – although questionable as clinically relevant – overall increase in mild and surgical complications. This is mostly caused by superficial wound infections when residents participate in surgical procedures," said Dr. P. Ravi Kiran, staff surgeon and head of the research section in the department of colorectal surgery at the Cleveland Clinic.

Using data from the National Surgical Quality Improvement Program database from 2005 to 2007, Dr. Kiran and his colleagues compared outcomes for patients who underwent surgery with and without resident participation.

The database, which includes data from pre-, intra-, and postoperative phases, uses clearly defined parameters and specialist nurse reviewers. It also includes resident participation and a morbidity probability, which offers an opportunity to use preoperative factors to stratify risk within subgroups, Dr. Kiran said at the annual meeting of the American Surgical Association.

Resident cases were matched with nonresident cases on the basis of age, sex, specialty, surgical procedure, morbidity probability, and important comorbidities and risk factors. Primary outcomes included 30-day mortality and postoperative complications (mild vs. severe, and surgical vs. medical). Secondary outcomes included the duration of surgery and length of hospital stay.

Mild complications included superficial surgical site infections (SSIs), peripheral nerve injury, urinary tract infection, deep venous thrombosis, and thrombophlebitis. Severe complications included deep (organ) SSI, wound disruption, bleeding requiring transfusion, failure of graft or prosthesis, reoperation, pneumonia, pulmonary embolism, acute renal failure, stroke, myocardial infarction, and sepsis.

Surgical complications included superficial SSI, deep (organ) SSI, wound disruption, bleeding requiring transfusion, failure of graft or prosthesis, peripheral nerve injury, and reoperation. Medical complications included pneumonia, pulmonary embolism, acute renal failure, stroke, myocardial infarction, sepsis, urinary tract infection, deep vein thrombosis, and thrombophlebitis.

For cases with resident vs. nonresident participation, the surgical complication rates were 7% and 6.2%, respectively – a significant difference – and mild complications rates were 4.4% and 3.5%, respectively. In addition, the mean operative time was significantly greater for cases involving residents – 122 vs. 97 minutes. The length of postoperative hospital stay was not significantly longer in the resident group.

The researchers identified 40,474 patients in the resident group and 20,237 patients in the nonresident group. The two groups were similar in terms of median age (50 years), sex (67% female), mean morbidity probability (0.09), American Society of Anesthesiologists classification, and presence of diabetes (6.4%) and hypertension (35%).

The groups were also similar in terms of presence of chronic obstructive pulmonary disease (0.27%), congestive heart failure or myocardial infarction in the past 6 months (0%), dialysis (0.044%), and preoperative sepsis (0.035%). Surgeons’ speciality areas were likewise similar for the two groups (general, 93%; vascular, 6%; and other, 1.36%).

Postgraduate year (PGY) 1-2 residents participated in 31% of operations, PGY 3-5 residents participated in 56%, and residents in PGY 6 or higher participated in 13% of cases.

The 10 most common surgical procedures were laparoscopic appendectomy, laparoscopic gastric bypass, laparoscopic cholecystectomy with and without operative cholangiogram, open appendectomy (nonruptured), thromboendarterectomy, colectomy (partial with anastomosis), laparoscopic colectomy (partial with anastomosis), ventral hernia repair, and placement of gastric band. These procedures were similar in terms of the percentages of resident and nonresident participation.

"We found that there was no difference in the [overall] 30-day mortality between the groups – 0.18% in the resident group and 0.20% in the no-resident group," said Dr. Kiran. However, any 30-day complications were 7.5% in the resident group and 6.7% in the nonresident group, a significant difference.

"When we further looked at the surgical complications, we noted that the cause of the difference in surgical complications between the two groups was the higher rate of SSIs in the resident group, when compared with the no-resident group ... the other surgical complications were similar," he said. The SSI rate was 3.0% for the resident group, compared with 2.2% for the nonresident group.

Interestingly, the researchers also found that overall 30-day complication rates increased with PGY – the rates were 6% for PGY 1-2, 8% for PGY 3-5, and 9% for PGY of 6 or more.

When they examined specific outcomes and complications between different PGY groups and matched cases without the involvement of residents, they found a similar pattern for the overall cohort.

"The reason for the difference in 30-day complications in the groups was because of differences in complications that were classified as mild, and primarily because the superficial surgical site infections were higher in the PGY 1-2 years, with an increased operative time," they said. The same was true for PGY 3-5 and PGY 6 and greater.

Also, as PGY increased, so did operative time – in both resident and nonresident groups. "This suggests that the reason for the increasing complications with increasing PGY years may have been related to increasing complexity of surgery," said Dr. Kiran.

"One overarching issue seems to be how we might achieve high-quality patient care and delivery of the clinical outcomes in the context of training," said Dr. Clifford Ko, a discussant.

However, he also acknowledged that teaching residents takes time. Dr. Ko, a colorectal surgeon and the director of the Center for Surgical Outcomes and Quality at the University of California, Los Angeles, questioned whether the longer operating time associated with resident involvement should be reduced.

"Although we would not perhaps be able to minimize time differences, I think that we have already achieved some mark of control by the gradation of responsibility over time, as residents continue with their training," Dr. Kiran said.

Although the surgical and mild complication rates were slightly greater, it’s unclear whether these differences are clinically relevant.

"The reasons for [these differences] are likely multifactorial and may be related to prolonged operative time. Considering that more complex cases may be performed in teaching hospitals and require resident participation, ‘resident’ could be a surrogate of severity of disease and intensity of operation – factors that may not be clearly discernible in a retrospective study – and this may explain the differences seen.

"Also, quality measures currently underway to reduce surgical site infections across the board may further minimize any of these differences that may exist," Dr. Kiran concluded.

The authors reported that they had no relevant disclosures.