- Email: [email protected]
Laparoscopic pyloromyotomy: effect of resident training on complications
Journal of Pediatric Surgery (2008) 43, 97–101
www.elsevier.com/locate/jpedsurg
Laparoscopic pyloromyotomy: effect of resident training on complications Ramanath N. Haricharan, Charles J. Aprahamian, Ahmet Celik, Carroll M. Harmon, Keith E. Georgeson, Douglas C. Barnhart⁎ Division of Pediatric Surgery, University of Alabama at Birmingham, Birmingham, AL 35233, USA Received 28 August 2007; accepted 2 September 2007
Index words: Hypertrophic pyloric stenosis; Laparoscopic surgery; Outcome measure; Complications; Academic training; Children
Abstract Purpose: The purpose of this study was to characterize the safety of laparoscopic pyloromyotomy and examine the effect of resident training on the occurrence of complications. Methods: Five hundred consecutive infants who underwent laparoscopic pyloromyotomy between January 1997 and December 2005 were reviewed and analyzed. Results: Laparoscopic pyloromyotomy was successfully completed in 489 patients (97.8%). Four hundred seventeen patients were boys (83%). Intraoperative complication occurred in 8 (1.6%) patients (mucosal perforation, 7; serosal injury to the duodenum, 1). All were immediately recognized and uneventfully repaired. Six patients (1.2%) required revision pyloromyotomy for persistent or recurrent gastric outlet obstruction. There were 7 wound complications (1.4%) and no deaths. Pediatric surgery residents performed 81% of the operations, whereas 16% were done by general surgery residents (postgraduate years 3-4). There was a 5.4-fold increased risk of mucosal perforation or incomplete pyloromyotomy when a general surgery resident rather than a pediatric surgery resident performed the operation (95% confidence interval, 1.8-15.8; P = .003). These effects persisted even after controlling for weight, age, and attending experience. Conclusions: The laparoscopic pyloromyotomy has an excellent success rate with low morbidity. The occurrence of complications is increased when the operation is performed by a general surgery resident, even when directly supervised by pediatric surgical faculty. © 2008 Elsevier Inc. All rights reserved.
Infantile hypertrophic pyloric stenosis is a common pediatric condition requiring surgical management. Extramucosal pyloromyotomy as described by Ramstedt via a right upper quadrant incision was the procedure of choice for decades [1,2]. Over the last decade, laparoscopic pyloro-
Presented at the 38th annual meeting of the American Pediatric Surgical Association, Orlando, Florida, May 24-27, 2007. ⁎ Corresponding author. Tel.: +1 205 939 9688; fax: +1 205 975 4972. E-mail address: [email protected] (D.C. Barnhart). 0022-3468/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2007.09.028
myotomy introduced by Alain et al [3] has gained popularity [1,4,5]. Several studies comparing open and laparoscopic pyloromyotomies have found overall similar complication rates [1,2,6-8], although the laparoscopic technique had better cosmetic results [1,2,9]. Others have suggested that the provider and hospital characteristics may affect the outcomes after a pyloromyotomy [10-13], irrespective of the approach. These authors suggest that better outcomes after pyloromyotomy are obtained at high-volume institutions [10,12] or by high-volume surgeons [10,12] or by surgeons with pediatric surgical training [11,13]. These studies, however, did not
98 estimate the possible effect of resident training on the outcome after pyloromyotomy. The purpose of this study was to review our experience with laparoscopic pyloromyotomy and analyze the effect of resident training on complications in a high-volume pediatric surgical training program.
1. Methods Five hundred consecutive infants who underwent laparoscopic pyloromyotomy at the Children's Hospital of Alabama, Birmingham, between January 1997 and December 2005 were included in the study. Approval by the institutional review board at University of Alabama at Birmingham was obtained before the review of the medical records. Data collected included demographics of the children, operative time, resident level (pediatric surgery or general surgery), pediatric surgery resident experience, attending pediatric surgeon experience, and intraoperative and postoperative complications including early (30 days) postoperative readmissions. Duodenal injury and mucosal perforation were regarded as intraoperative complications, and revision pyloromyotomy for persistent gastric outlet obstruction, readmission for persistent vomiting and wound complications were considered as postoperative complications. Duodenal injury, mucosal perforation, and revision pyloromyotomy were combined to estimate the techniquerelated complication rate. Laparoscopic pyloromyotomy was performed by either a general surgery resident (postgraduate year [PGY] 3 or PGY4) or a pediatric surgery resident (fellow) with direct and constant supervision by an attending pediatric surgeon. Attending surgeon experience for each case was indicated by the case order for the particular surgeon from the beginning of the study period. For example, an attending supervising his 10th case during the study period was considered to have an experience of 10 and the same attending supervising his 25th case during the study period was considered to have an experience of 25. Similar methodology was used to assign the experience level for each pediatric surgery resident during the study period. General surgery resident participation in the procedure was analyzed as a dichotomous variable. Postoperatively, the feeding regimen was independent of the level of the resident involved intraoperatively. The attending pediatric surgeons polled before analyzing the study suggested that the pediatric surgery residents had to perform 15 pyloromyotomies before they would consider the residents proficient in the procedure. This cut-point was used to compare the occurrence of complications in the initial vs later experience of pediatric surgery residents. Statistical analyses were performed using Fisher's Exact test and logistic regression using the SAS 9.1 software program (SAS Institute, Cary, NC). A P-value of less than .05 was considered significant by convention.
R.N. Haricharan et al.
2. Results 2.1. Study population Five hundred children undergoing laparoscopic pyloromyotomies were analyzed. Characteristics of the children are as shown in Table 1. There were a total of 8 attending pediatric surgeons and 13 pediatric surgery residents (fellows) that contributed to the laparoscopic pyloromyotomies included in the study. The median number of cases by an attending surgeon was 52 (range, 15-173 cases) and the median number of cases by a pediatric surgery resident was 31 (range, 5-60 cases). There was no significant difference in the age, the intraoperative weight, or the operative time between the children operated by pediatric surgery residents and those by general surgery residents.
2.2. Operative time The overall mean operative time was 26 minutes (SD, 11minutes). As shown in Fig. 1, the operative time did not differ significantly between general surgery residents (28 ± 11 minutes) and pediatric surgery residents (26 ± 11 minutes). Among pediatric surgery residents, the experience level was grouped into 4 groups. There was no significant difference in their operative times as their case experience increased, although the variance decreased as they progressed to higher-experience groups.
2.3. Complications Table 2 shows the distribution of technique-related complications (duodenal injury, mucosal perforations, and revision pyloromyotomy) by the level of resident involved in the case. Pediatric surgery residents performed most (81%) of the pyloromyotomies. Fourteen cases were done by the attending pediatric surgeon without resident involvement, and there were no complications in these cases. The overall rate of perforation was 1.4% (95% confidence interval [CI], 0.4-2.4) and the rate of revision pyloromyotomy was 1.2% (95% CI, 0.2-2.2). All mucosal perforations and the duodenal injury were immediately recognized in the operating room and were uneventfully repaired. Persistent gastric outlet obstruction seen in 6 children was confirmed
Table 1
Study population details
Total N Laparoscopic pyloromyotomy completed successfully Males (n) Age at operation (median, wk) Weight (median, kg) Operating time (mean, min)
500 489
97.8%
417 5 3.8 26
83.4% range, 1-19 range, 2.0-6.2 ±11 SD
Laparoscopic pyloromyotomy resident training effect
Fig. 1
Operative time by resident experience level.
with an upper gastrointestinal study (5 children) or ultrasound (1 child) before taking them to the operating room for revision pyloromyotomy. Intraoperatively, it was seen that all 6 children had fused the edges of their previous myotomy site. The procedure was done after a median interval of 3 weeks (range, 2-5 weeks) after initial procedure; 4 of 6 were done by the open approach and 2 were done laparoscopically. None of these children required a second revision procedure. Eleven (2.2%) children underwent conversions to an open pyloromyotomy, 7 (64%) of those were secondary to mucosal perforation; other reasons were inadequate visualization (2 children), duodenal injury (1 child), or inadequate pyloromyotomy (1 child). There were no deaths in the study population. The complication rates did not differ significantly between the attending pediatric surgeons. The general surgery residents had a 5.4 times (95% CI, 1.8-15.8; P = .003) risk of a technique-related complication when compared to the pediatric surgery residents. Further analyses of individual technique-related complications between the 2 groups of residents showed a significant increase in risk of mucosal perforation (P = .01) when a general surgery resident was the trainee resident. The trend of increased risk of duodenal injury and revision pyloromyotomy failed to reach statistical significance. Because most of Table 2
the conversions to a laparotomy were done secondary to intraoperative complications, the risk of conversion was lower when the operation was performed by a pediatric surgery resident as compared to a general surgery resident (odds ratio, 0.23; 95% CI, 0.07-0.76; P = .02). These effects persisted even after the effects of intraoperative weight, age at operation, and attending experience were controlled. The occurrence of technique-related complications by pediatric surgery resident experience is shown in Table 3. There was no significant difference in the complication rate between the groups, although there was a steady decline in the incidence after the first 15 cases. The port-site complications occurred in 7 (1.4%) children and included omental herniation requiring operation in 4, wound infection in 2, and suture granuloma requiring excision in 1 child. All 11 children (2.2%) with readmissions within 30 days postoperatively had persistent vomiting. Persistent or recurrent gastric outlet obstruction was ruled out in these children by further radiologic investigation. Five of these children were diagnosed to have gastroesophageal reflux, 3 had gastroenteritis, and 3 had other nonobstructive causes of vomiting. The median interval between pyloromyotomy and readmission was 1 week (range, 2 days5 weeks) and their median hospital stay was 2 days (range, 1-6 days).
Technique-related complications in laparoscopic pyloromyotomy
Pyloromyotomies Conversions Technique-related Total Mucosal perforation Duodenal injury Revision pyloromyotomy a
99
No. of cases by residents a
Total n (%)
Pediatric surgery
General surgery
500 11 (2.2%)
406 6 (1.4%)
80 5 (6.3%)
14 (2.8%) 7 (1.4%) 1 (0.2%) 6 (1.2%)
7 3 0 4
7 4 1 2
Fourteen cases were done by the attending pediatric surgeons without a resident.
Odds ratio (95%CI)
P
4.3 (1.3-14.3)
.02
5.4 (1.8-15.8) 6.7 (1.5-29.6) 15.0 (0.6-366.0) 2.5 (0.46-13.6)
.003 .01 .16 .25
100
R.N. Haricharan et al.
Table 3 Technique-related complications by pediatric surgery resident experience Experience category
n
Technique-related complications
1-15 cases 16-30 cases 31-45 cases 46-60 cases
179 124 69 33
4 2 1 0
(2.2%) (1.6%) (1.4%) (0%)
The group of initial 15 operations performed by each pediatric surgery resident was compared to the group of cases by general surgery residents and to the group of all the subsequent cases by pediatric surgery residents to better characterize a training effect. The general surgery residents had a 4.1 times higher risk (95% CI, 1.2-14.6; P = .02) of technical complications compared to the risk in the initial operations done by the pediatric surgery residents. There was no significant increased risk associated with the initial 15 operations compared with the subsequent cases performed by each pediatric surgery resident (odds ratio, 1.7; P = .37). These training effects did not change after controlling for the effects of weight, age, and attending experience.
3. Discussion This study presents the largest single-institution series of laparoscopic pyloromyotomy to date. Overall, the current series shows that laparoscopic pyloromyotomy has an excellent successful completion rate (97.8%) with low postoperative complication rate (4.8%) including minor wound complications and readmissions for vomiting (of nonobstructive etiology). The rate of perforation during a laparoscopic pyloromyotomy reported in the literature ranges from 0% [1] to 9% [5], and the incidence of revision pyloromyotomy ranges from 0% [1] to 5.5% [14]. The rate of perforation (1.4%) and the rate of revision pyloromyotomy (1.2%) estimated in this study are toward the lower end of the rates reported in the literature, likely reflecting the lack of the attending surgeon learning curve phenomenon in this series. The estimated rates are comparable to the complication rates after open pyloromyotomy (rate of perforation, 0% [15] to 5.9% [5]; and rate of revision pyloromyotomy, 0% [2] to 2% [5]). Several articles comparing the laparoscopic and open operations concluded that the laparoscopic technique takes a longer duration to perform and has a higher complication rate [14,16]. Our study does not support such conclusions. As the authors acknowledged in these articles, the initial learning curve may have accounted for those findings. It was interesting to note that the profile of complications after laparoscopic pyloromyotomy at our institution has changed over the years. In a previous series from our institution [2], we had reported an incomplete pyloromyotomy rate of 2.2% and a
mucosal perforation rate of 0.4% after a laparoscopic approach, which was possibly because of the decreased aggressiveness of the surgeons earlier in the experience. Whereas in the current study, we see a trend in the opposite direction with an incomplete pyloromyotomy rate of 1.2% and a mucosal perforation rate of 1.4%, which may be attributed to the attempts at rectifying the problem with incomplete pyloromyotomies. Given that laparoscopic pyloromyotomy is usually performed in infants and is a procedure that requires advanced laparoscopic skills, the authors hypothesized that a resident training effect might influence the occurrence of complications. A high-volume institution like ours is an excellent setup to study the effects of resident training, as the major factors influencing the outcome suggested by other authors (hospital volume, surgical volume, and specialized pediatric care) are controlled. Prior studies evaluating the influence of resident level on operative outcomes have been variable [17-22]. Resident training effects on the outcomes have been reported in adult surgery literature for laparoscopic cholecystectomy [20,23] and inguinal hernia repair [22]. Studies showed a higher rate of conversion to laparotomy [23] and an increased complication rate [20] of laparoscopic cholecystectomy with decrease in resident postgraduate year. Most of these studies showed an increased cost with junior residents when compared with senior residents in the form of increased operative times [20,22]. Several studies have shown that operating with residents in a teaching program per se does not have any negative effect on outcomes when compared to procedures performed by attending surgeons in nonteaching institutions [19,21,24]. There is a repeated learning curve associated with the general surgery residents performing pediatric laparoscopic surgery that returns to the initial baseline with each resident completing the rotation, contributing to the resident training effect. In fact, despite low overall technique-related complications, patients who underwent operation with a general surgery resident as the trainee had a higher risk when compared to the pediatric surgery residents. The general surgery residents had a 5-fold higher risk of a techniquerelated complication when compared to the pediatric surgery residents. A similar effect persisted even when compared to the initial experience of the pediatric surgery residents. Among the pediatric surgery residents, there was no significant difference in the complications between the initial 15 and the subsequent operations. Although the incidence of complications reduced as they completed the initial 15 operations, we failed to see a significant effect possibly because of insufficient power to detect differences in such a low-complication situation. In patients with intraoperative complications (duodenal injury or mucosal perforation), the consequence was a conversion to an open operation. As discussed earlier, it has been shown that there is no additional morbidity to the open procedure when compared to the laparoscopic procedure except for a less aesthetic postoperative scar. In patients with
Laparoscopic pyloromyotomy resident training effect an incomplete pyloromyotomy, the consequence was an additional hospital admission and a revision pyloromyotomy, and this was not significantly different between the 2 resident groups. It is essential to note that although these events are classified as complications, 8 of 14 resulted only in conversion to an open pyloromyotomy and the other 6 resulted in repeat admission with otherwise uneventful revision pyloromyotomy. Our study differs from the study by Gollin et al [18] where they analyzed resident level and postoperative outcomes after several pediatric laparoscopic surgeries including pyloromyotomy. They did not see any difference in outcomes upon comparing PGY2, PGY3, and PGY4 residents with technical complication rates of 10% (1/10), 0 (0/24), and 2.6% (1/38), respectively. Because of the overall low technical complication rate (2.8%) in their series, the sample size may have been inadequate to detect a clinically important difference. Furthermore, their study did not compare the general surgery resident outcomes to pediatric surgery resident outcomes (control). A key factor in the pyloromyotomy is the appropriate pressure applied over the hypertrophied pylorus, which directly influences the outcome of the procedure. The feedback provided by the sensation of resistance by the pylorus is hard to replicate and difficult for the attending surgeon to demonstrate and supervise. The laparoscopic pyloromyotomy may be a unique procedure because of the dependence on tactile feedback to safely complete the operation. This factor may limit the generalizability of our study to other procedures. This study does raise the concern that certain procedures may be associated with increased risk to the patient when performed by more junior trainees. Surgeons involved in training residents must consider whether this increased risk is ethical especially if the resident is not likely to perform the procedure after the completion of the training. The laparoscopic pyloromyotomy has an excellent success rate with low morbidity. The risk of techniquerelated complications, although very low, is increased when the operation is performed by a general surgery resident even when directly supervised by pediatric surgical faculty.
References [1] St Peter SD, Holcomb III GW, Calkins CM, et al. Open versus laparoscopic pyloromyotomy for pyloric stenosis: a prospective, randomized trial. Ann Surg 2006;244:363-70. [2] Yagmurlu A, Barnhart DC, Vernon A, et al. Comparison of the incidence of complications in open and laparoscopic pyloromyotomy: a concurrent single institution series. J Pediatr Surg 2004;39:292-6 [discussion 292-296].
101 [3] Alain JL, Grousseau D, Terrier G. Extramucosal pyloromyotomy by laparoscopy. Surg Endosc 1991;5:174-5. [4] Kim SS, Lau ST, Lee SL, et al. The learning curve associated with laparoscopic pyloromyotomy. J Laparoendosc Adv Surg Tech A 2005; 15:474-7. [5] Ford WD, Crameri JA, Holland AJ. The learning curve for laparoscopic pyloromyotomy. J Pediatr Surg 1997;32:552-4. [6] Adibe OO, Nichol PF, Flake AW, et al. Comparison of outcomes after laparoscopic and open pyloromyotomy at a high-volume pediatric teaching hospital. J Pediatr Surg 2006;41:1676-8. [7] Hall NJ, Ade-Ajayi N, Al-Roubaie J, et al. Retrospective comparison of open versus laparoscopic pyloromyotomy. Br J Surg 2004;91: 1325-9. [8] Kim SS, Lau ST, Lee SL, et al. Pyloromyotomy: a comparison of laparoscopic, circumumbilical, and right upper quadrant operative techniques. J Am Coll Surg 2005;201:66-70. [9] Hall NJ, Van Der Zee J, Tan HL, et al. Meta-analysis of laparoscopic versus open pyloromyotomy. Ann Surg 2004;240:774-8. [10] Safford SD, Pietrobon R, Safford KM, et al. A study of 11,003 patients with hypertrophic pyloric stenosis and the association between surgeon and hospital volume and outcomes. J Pediatr Surg 2005;40:967-72 [discussion 972-963]. [11] Allan C. Determinants of good outcome in pyloric stenosis. J Paediatr Child Health 2006;42:86-8. [12] Ly DP, Liao JG, Burd RS. Effect of surgeon and hospital characteristics on outcome after pyloromyotomy. Arch Surg 2005;140:1191-7. [13] Pranikoff T, Campbell BT, Travis J, et al. Differences in outcome with subspecialty care: pyloromyotomy in North Carolina. J Pediatr Surg 2002;37:352-6. [14] Sitsen E, Bax NM, van der Zee DC. Is laparoscopic pyloromyotomy superior to open surgery? Surg Endosc 1998;12:813-5. [15] Bufo AJ, Merry C, Shah R, et al. Laparoscopic pyloromyotomy: a safer technique. Pediatr Surg Int 1998;13:240-2. [16] Campbell BT, McLean K, Barnhart DC, et al. A comparison of laparoscopic and open pyloromyotomy at a teaching hospital. J Pediatr Surg 2002;37:1068-71 [discussion 1068-1071]. [17] Wang WN, Melkonian MG, Marshall R, et al. Postgraduate year does not influence operating time in laparoscopic cholecystectomy. J Surg Res 2001;101:1-3. [18] Gollin G, Moores D, Baerg JC. Getting residents in the game: an evaluation of general surgery residents' participation in pediatric laparoscopic surgery. J Pediatr Surg 2004;39:78-80. [19] Hutter MM, Glasgow RE, Mulvihill SJ. Does the participation of a surgical trainee adversely impact patient outcomes? A study of major pancreatic resections in California. Surgery 2000;128:286-92. [20] Kauvar DS, Braswell A, Brown BD, et al. Influence of resident and attending surgeon seniority on operative performance in laparoscopic cholecystectomy. J Surg Res 2006;132:159-63. [21] Praseedom RK, Paisley A, Madhavan KK, et al. Supervised surgical trainees can perform pancreatic resections safely. J R Coll Surg Edinb 1999;44:16-8. [22] Wilkiemeyer M, Pappas TN, Giobbie-Hurder A, et al. Does resident post graduate year influence the outcomes of inguinal hernia repair? Ann Surg 2005;241:879-82 [discussion 882-874]. [23] Bickel A, Rappaport A, Hazani E, et al. Laparoscopic cholecystectomy for acute cholecystitis performed by residents in surgery: a risk factor for conversion to open laparotomy? J Laparoendosc Adv Surg Tech A 1998;8:137-41. [24] Bottsford JH, Blake R, Small T, et al. Carotid endarterectomy: the outcome is not adversely affected by surgical residents. J S C Med Assoc 1997;93:243-7.
www.elsevier.com/locate/jpedsurg
Laparoscopic pyloromyotomy: effect of resident training on complications Ramanath N. Haricharan, Charles J. Aprahamian, Ahmet Celik, Carroll M. Harmon, Keith E. Georgeson, Douglas C. Barnhart⁎ Division of Pediatric Surgery, University of Alabama at Birmingham, Birmingham, AL 35233, USA Received 28 August 2007; accepted 2 September 2007
Index words: Hypertrophic pyloric stenosis; Laparoscopic surgery; Outcome measure; Complications; Academic training; Children
Abstract Purpose: The purpose of this study was to characterize the safety of laparoscopic pyloromyotomy and examine the effect of resident training on the occurrence of complications. Methods: Five hundred consecutive infants who underwent laparoscopic pyloromyotomy between January 1997 and December 2005 were reviewed and analyzed. Results: Laparoscopic pyloromyotomy was successfully completed in 489 patients (97.8%). Four hundred seventeen patients were boys (83%). Intraoperative complication occurred in 8 (1.6%) patients (mucosal perforation, 7; serosal injury to the duodenum, 1). All were immediately recognized and uneventfully repaired. Six patients (1.2%) required revision pyloromyotomy for persistent or recurrent gastric outlet obstruction. There were 7 wound complications (1.4%) and no deaths. Pediatric surgery residents performed 81% of the operations, whereas 16% were done by general surgery residents (postgraduate years 3-4). There was a 5.4-fold increased risk of mucosal perforation or incomplete pyloromyotomy when a general surgery resident rather than a pediatric surgery resident performed the operation (95% confidence interval, 1.8-15.8; P = .003). These effects persisted even after controlling for weight, age, and attending experience. Conclusions: The laparoscopic pyloromyotomy has an excellent success rate with low morbidity. The occurrence of complications is increased when the operation is performed by a general surgery resident, even when directly supervised by pediatric surgical faculty. © 2008 Elsevier Inc. All rights reserved.
Infantile hypertrophic pyloric stenosis is a common pediatric condition requiring surgical management. Extramucosal pyloromyotomy as described by Ramstedt via a right upper quadrant incision was the procedure of choice for decades [1,2]. Over the last decade, laparoscopic pyloro-
Presented at the 38th annual meeting of the American Pediatric Surgical Association, Orlando, Florida, May 24-27, 2007. ⁎ Corresponding author. Tel.: +1 205 939 9688; fax: +1 205 975 4972. E-mail address: [email protected] (D.C. Barnhart). 0022-3468/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2007.09.028
myotomy introduced by Alain et al [3] has gained popularity [1,4,5]. Several studies comparing open and laparoscopic pyloromyotomies have found overall similar complication rates [1,2,6-8], although the laparoscopic technique had better cosmetic results [1,2,9]. Others have suggested that the provider and hospital characteristics may affect the outcomes after a pyloromyotomy [10-13], irrespective of the approach. These authors suggest that better outcomes after pyloromyotomy are obtained at high-volume institutions [10,12] or by high-volume surgeons [10,12] or by surgeons with pediatric surgical training [11,13]. These studies, however, did not
98 estimate the possible effect of resident training on the outcome after pyloromyotomy. The purpose of this study was to review our experience with laparoscopic pyloromyotomy and analyze the effect of resident training on complications in a high-volume pediatric surgical training program.
1. Methods Five hundred consecutive infants who underwent laparoscopic pyloromyotomy at the Children's Hospital of Alabama, Birmingham, between January 1997 and December 2005 were included in the study. Approval by the institutional review board at University of Alabama at Birmingham was obtained before the review of the medical records. Data collected included demographics of the children, operative time, resident level (pediatric surgery or general surgery), pediatric surgery resident experience, attending pediatric surgeon experience, and intraoperative and postoperative complications including early (30 days) postoperative readmissions. Duodenal injury and mucosal perforation were regarded as intraoperative complications, and revision pyloromyotomy for persistent gastric outlet obstruction, readmission for persistent vomiting and wound complications were considered as postoperative complications. Duodenal injury, mucosal perforation, and revision pyloromyotomy were combined to estimate the techniquerelated complication rate. Laparoscopic pyloromyotomy was performed by either a general surgery resident (postgraduate year [PGY] 3 or PGY4) or a pediatric surgery resident (fellow) with direct and constant supervision by an attending pediatric surgeon. Attending surgeon experience for each case was indicated by the case order for the particular surgeon from the beginning of the study period. For example, an attending supervising his 10th case during the study period was considered to have an experience of 10 and the same attending supervising his 25th case during the study period was considered to have an experience of 25. Similar methodology was used to assign the experience level for each pediatric surgery resident during the study period. General surgery resident participation in the procedure was analyzed as a dichotomous variable. Postoperatively, the feeding regimen was independent of the level of the resident involved intraoperatively. The attending pediatric surgeons polled before analyzing the study suggested that the pediatric surgery residents had to perform 15 pyloromyotomies before they would consider the residents proficient in the procedure. This cut-point was used to compare the occurrence of complications in the initial vs later experience of pediatric surgery residents. Statistical analyses were performed using Fisher's Exact test and logistic regression using the SAS 9.1 software program (SAS Institute, Cary, NC). A P-value of less than .05 was considered significant by convention.
R.N. Haricharan et al.
2. Results 2.1. Study population Five hundred children undergoing laparoscopic pyloromyotomies were analyzed. Characteristics of the children are as shown in Table 1. There were a total of 8 attending pediatric surgeons and 13 pediatric surgery residents (fellows) that contributed to the laparoscopic pyloromyotomies included in the study. The median number of cases by an attending surgeon was 52 (range, 15-173 cases) and the median number of cases by a pediatric surgery resident was 31 (range, 5-60 cases). There was no significant difference in the age, the intraoperative weight, or the operative time between the children operated by pediatric surgery residents and those by general surgery residents.
2.2. Operative time The overall mean operative time was 26 minutes (SD, 11minutes). As shown in Fig. 1, the operative time did not differ significantly between general surgery residents (28 ± 11 minutes) and pediatric surgery residents (26 ± 11 minutes). Among pediatric surgery residents, the experience level was grouped into 4 groups. There was no significant difference in their operative times as their case experience increased, although the variance decreased as they progressed to higher-experience groups.
2.3. Complications Table 2 shows the distribution of technique-related complications (duodenal injury, mucosal perforations, and revision pyloromyotomy) by the level of resident involved in the case. Pediatric surgery residents performed most (81%) of the pyloromyotomies. Fourteen cases were done by the attending pediatric surgeon without resident involvement, and there were no complications in these cases. The overall rate of perforation was 1.4% (95% confidence interval [CI], 0.4-2.4) and the rate of revision pyloromyotomy was 1.2% (95% CI, 0.2-2.2). All mucosal perforations and the duodenal injury were immediately recognized in the operating room and were uneventfully repaired. Persistent gastric outlet obstruction seen in 6 children was confirmed
Table 1
Study population details
Total N Laparoscopic pyloromyotomy completed successfully Males (n) Age at operation (median, wk) Weight (median, kg) Operating time (mean, min)
500 489
97.8%
417 5 3.8 26
83.4% range, 1-19 range, 2.0-6.2 ±11 SD
Laparoscopic pyloromyotomy resident training effect
Fig. 1
Operative time by resident experience level.
with an upper gastrointestinal study (5 children) or ultrasound (1 child) before taking them to the operating room for revision pyloromyotomy. Intraoperatively, it was seen that all 6 children had fused the edges of their previous myotomy site. The procedure was done after a median interval of 3 weeks (range, 2-5 weeks) after initial procedure; 4 of 6 were done by the open approach and 2 were done laparoscopically. None of these children required a second revision procedure. Eleven (2.2%) children underwent conversions to an open pyloromyotomy, 7 (64%) of those were secondary to mucosal perforation; other reasons were inadequate visualization (2 children), duodenal injury (1 child), or inadequate pyloromyotomy (1 child). There were no deaths in the study population. The complication rates did not differ significantly between the attending pediatric surgeons. The general surgery residents had a 5.4 times (95% CI, 1.8-15.8; P = .003) risk of a technique-related complication when compared to the pediatric surgery residents. Further analyses of individual technique-related complications between the 2 groups of residents showed a significant increase in risk of mucosal perforation (P = .01) when a general surgery resident was the trainee resident. The trend of increased risk of duodenal injury and revision pyloromyotomy failed to reach statistical significance. Because most of Table 2
the conversions to a laparotomy were done secondary to intraoperative complications, the risk of conversion was lower when the operation was performed by a pediatric surgery resident as compared to a general surgery resident (odds ratio, 0.23; 95% CI, 0.07-0.76; P = .02). These effects persisted even after the effects of intraoperative weight, age at operation, and attending experience were controlled. The occurrence of technique-related complications by pediatric surgery resident experience is shown in Table 3. There was no significant difference in the complication rate between the groups, although there was a steady decline in the incidence after the first 15 cases. The port-site complications occurred in 7 (1.4%) children and included omental herniation requiring operation in 4, wound infection in 2, and suture granuloma requiring excision in 1 child. All 11 children (2.2%) with readmissions within 30 days postoperatively had persistent vomiting. Persistent or recurrent gastric outlet obstruction was ruled out in these children by further radiologic investigation. Five of these children were diagnosed to have gastroesophageal reflux, 3 had gastroenteritis, and 3 had other nonobstructive causes of vomiting. The median interval between pyloromyotomy and readmission was 1 week (range, 2 days5 weeks) and their median hospital stay was 2 days (range, 1-6 days).
Technique-related complications in laparoscopic pyloromyotomy
Pyloromyotomies Conversions Technique-related Total Mucosal perforation Duodenal injury Revision pyloromyotomy a
99
No. of cases by residents a
Total n (%)
Pediatric surgery
General surgery
500 11 (2.2%)
406 6 (1.4%)
80 5 (6.3%)
14 (2.8%) 7 (1.4%) 1 (0.2%) 6 (1.2%)
7 3 0 4
7 4 1 2
Fourteen cases were done by the attending pediatric surgeons without a resident.
Odds ratio (95%CI)
P
4.3 (1.3-14.3)
.02
5.4 (1.8-15.8) 6.7 (1.5-29.6) 15.0 (0.6-366.0) 2.5 (0.46-13.6)
.003 .01 .16 .25
100
R.N. Haricharan et al.
Table 3 Technique-related complications by pediatric surgery resident experience Experience category
n
Technique-related complications
1-15 cases 16-30 cases 31-45 cases 46-60 cases
179 124 69 33
4 2 1 0
(2.2%) (1.6%) (1.4%) (0%)
The group of initial 15 operations performed by each pediatric surgery resident was compared to the group of cases by general surgery residents and to the group of all the subsequent cases by pediatric surgery residents to better characterize a training effect. The general surgery residents had a 4.1 times higher risk (95% CI, 1.2-14.6; P = .02) of technical complications compared to the risk in the initial operations done by the pediatric surgery residents. There was no significant increased risk associated with the initial 15 operations compared with the subsequent cases performed by each pediatric surgery resident (odds ratio, 1.7; P = .37). These training effects did not change after controlling for the effects of weight, age, and attending experience.
3. Discussion This study presents the largest single-institution series of laparoscopic pyloromyotomy to date. Overall, the current series shows that laparoscopic pyloromyotomy has an excellent successful completion rate (97.8%) with low postoperative complication rate (4.8%) including minor wound complications and readmissions for vomiting (of nonobstructive etiology). The rate of perforation during a laparoscopic pyloromyotomy reported in the literature ranges from 0% [1] to 9% [5], and the incidence of revision pyloromyotomy ranges from 0% [1] to 5.5% [14]. The rate of perforation (1.4%) and the rate of revision pyloromyotomy (1.2%) estimated in this study are toward the lower end of the rates reported in the literature, likely reflecting the lack of the attending surgeon learning curve phenomenon in this series. The estimated rates are comparable to the complication rates after open pyloromyotomy (rate of perforation, 0% [15] to 5.9% [5]; and rate of revision pyloromyotomy, 0% [2] to 2% [5]). Several articles comparing the laparoscopic and open operations concluded that the laparoscopic technique takes a longer duration to perform and has a higher complication rate [14,16]. Our study does not support such conclusions. As the authors acknowledged in these articles, the initial learning curve may have accounted for those findings. It was interesting to note that the profile of complications after laparoscopic pyloromyotomy at our institution has changed over the years. In a previous series from our institution [2], we had reported an incomplete pyloromyotomy rate of 2.2% and a
mucosal perforation rate of 0.4% after a laparoscopic approach, which was possibly because of the decreased aggressiveness of the surgeons earlier in the experience. Whereas in the current study, we see a trend in the opposite direction with an incomplete pyloromyotomy rate of 1.2% and a mucosal perforation rate of 1.4%, which may be attributed to the attempts at rectifying the problem with incomplete pyloromyotomies. Given that laparoscopic pyloromyotomy is usually performed in infants and is a procedure that requires advanced laparoscopic skills, the authors hypothesized that a resident training effect might influence the occurrence of complications. A high-volume institution like ours is an excellent setup to study the effects of resident training, as the major factors influencing the outcome suggested by other authors (hospital volume, surgical volume, and specialized pediatric care) are controlled. Prior studies evaluating the influence of resident level on operative outcomes have been variable [17-22]. Resident training effects on the outcomes have been reported in adult surgery literature for laparoscopic cholecystectomy [20,23] and inguinal hernia repair [22]. Studies showed a higher rate of conversion to laparotomy [23] and an increased complication rate [20] of laparoscopic cholecystectomy with decrease in resident postgraduate year. Most of these studies showed an increased cost with junior residents when compared with senior residents in the form of increased operative times [20,22]. Several studies have shown that operating with residents in a teaching program per se does not have any negative effect on outcomes when compared to procedures performed by attending surgeons in nonteaching institutions [19,21,24]. There is a repeated learning curve associated with the general surgery residents performing pediatric laparoscopic surgery that returns to the initial baseline with each resident completing the rotation, contributing to the resident training effect. In fact, despite low overall technique-related complications, patients who underwent operation with a general surgery resident as the trainee had a higher risk when compared to the pediatric surgery residents. The general surgery residents had a 5-fold higher risk of a techniquerelated complication when compared to the pediatric surgery residents. A similar effect persisted even when compared to the initial experience of the pediatric surgery residents. Among the pediatric surgery residents, there was no significant difference in the complications between the initial 15 and the subsequent operations. Although the incidence of complications reduced as they completed the initial 15 operations, we failed to see a significant effect possibly because of insufficient power to detect differences in such a low-complication situation. In patients with intraoperative complications (duodenal injury or mucosal perforation), the consequence was a conversion to an open operation. As discussed earlier, it has been shown that there is no additional morbidity to the open procedure when compared to the laparoscopic procedure except for a less aesthetic postoperative scar. In patients with
Laparoscopic pyloromyotomy resident training effect an incomplete pyloromyotomy, the consequence was an additional hospital admission and a revision pyloromyotomy, and this was not significantly different between the 2 resident groups. It is essential to note that although these events are classified as complications, 8 of 14 resulted only in conversion to an open pyloromyotomy and the other 6 resulted in repeat admission with otherwise uneventful revision pyloromyotomy. Our study differs from the study by Gollin et al [18] where they analyzed resident level and postoperative outcomes after several pediatric laparoscopic surgeries including pyloromyotomy. They did not see any difference in outcomes upon comparing PGY2, PGY3, and PGY4 residents with technical complication rates of 10% (1/10), 0 (0/24), and 2.6% (1/38), respectively. Because of the overall low technical complication rate (2.8%) in their series, the sample size may have been inadequate to detect a clinically important difference. Furthermore, their study did not compare the general surgery resident outcomes to pediatric surgery resident outcomes (control). A key factor in the pyloromyotomy is the appropriate pressure applied over the hypertrophied pylorus, which directly influences the outcome of the procedure. The feedback provided by the sensation of resistance by the pylorus is hard to replicate and difficult for the attending surgeon to demonstrate and supervise. The laparoscopic pyloromyotomy may be a unique procedure because of the dependence on tactile feedback to safely complete the operation. This factor may limit the generalizability of our study to other procedures. This study does raise the concern that certain procedures may be associated with increased risk to the patient when performed by more junior trainees. Surgeons involved in training residents must consider whether this increased risk is ethical especially if the resident is not likely to perform the procedure after the completion of the training. The laparoscopic pyloromyotomy has an excellent success rate with low morbidity. The risk of techniquerelated complications, although very low, is increased when the operation is performed by a general surgery resident even when directly supervised by pediatric surgical faculty.
References [1] St Peter SD, Holcomb III GW, Calkins CM, et al. Open versus laparoscopic pyloromyotomy for pyloric stenosis: a prospective, randomized trial. Ann Surg 2006;244:363-70. [2] Yagmurlu A, Barnhart DC, Vernon A, et al. Comparison of the incidence of complications in open and laparoscopic pyloromyotomy: a concurrent single institution series. J Pediatr Surg 2004;39:292-6 [discussion 292-296].
101 [3] Alain JL, Grousseau D, Terrier G. Extramucosal pyloromyotomy by laparoscopy. Surg Endosc 1991;5:174-5. [4] Kim SS, Lau ST, Lee SL, et al. The learning curve associated with laparoscopic pyloromyotomy. J Laparoendosc Adv Surg Tech A 2005; 15:474-7. [5] Ford WD, Crameri JA, Holland AJ. The learning curve for laparoscopic pyloromyotomy. J Pediatr Surg 1997;32:552-4. [6] Adibe OO, Nichol PF, Flake AW, et al. Comparison of outcomes after laparoscopic and open pyloromyotomy at a high-volume pediatric teaching hospital. J Pediatr Surg 2006;41:1676-8. [7] Hall NJ, Ade-Ajayi N, Al-Roubaie J, et al. Retrospective comparison of open versus laparoscopic pyloromyotomy. Br J Surg 2004;91: 1325-9. [8] Kim SS, Lau ST, Lee SL, et al. Pyloromyotomy: a comparison of laparoscopic, circumumbilical, and right upper quadrant operative techniques. J Am Coll Surg 2005;201:66-70. [9] Hall NJ, Van Der Zee J, Tan HL, et al. Meta-analysis of laparoscopic versus open pyloromyotomy. Ann Surg 2004;240:774-8. [10] Safford SD, Pietrobon R, Safford KM, et al. A study of 11,003 patients with hypertrophic pyloric stenosis and the association between surgeon and hospital volume and outcomes. J Pediatr Surg 2005;40:967-72 [discussion 972-963]. [11] Allan C. Determinants of good outcome in pyloric stenosis. J Paediatr Child Health 2006;42:86-8. [12] Ly DP, Liao JG, Burd RS. Effect of surgeon and hospital characteristics on outcome after pyloromyotomy. Arch Surg 2005;140:1191-7. [13] Pranikoff T, Campbell BT, Travis J, et al. Differences in outcome with subspecialty care: pyloromyotomy in North Carolina. J Pediatr Surg 2002;37:352-6. [14] Sitsen E, Bax NM, van der Zee DC. Is laparoscopic pyloromyotomy superior to open surgery? Surg Endosc 1998;12:813-5. [15] Bufo AJ, Merry C, Shah R, et al. Laparoscopic pyloromyotomy: a safer technique. Pediatr Surg Int 1998;13:240-2. [16] Campbell BT, McLean K, Barnhart DC, et al. A comparison of laparoscopic and open pyloromyotomy at a teaching hospital. J Pediatr Surg 2002;37:1068-71 [discussion 1068-1071]. [17] Wang WN, Melkonian MG, Marshall R, et al. Postgraduate year does not influence operating time in laparoscopic cholecystectomy. J Surg Res 2001;101:1-3. [18] Gollin G, Moores D, Baerg JC. Getting residents in the game: an evaluation of general surgery residents' participation in pediatric laparoscopic surgery. J Pediatr Surg 2004;39:78-80. [19] Hutter MM, Glasgow RE, Mulvihill SJ. Does the participation of a surgical trainee adversely impact patient outcomes? A study of major pancreatic resections in California. Surgery 2000;128:286-92. [20] Kauvar DS, Braswell A, Brown BD, et al. Influence of resident and attending surgeon seniority on operative performance in laparoscopic cholecystectomy. J Surg Res 2006;132:159-63. [21] Praseedom RK, Paisley A, Madhavan KK, et al. Supervised surgical trainees can perform pancreatic resections safely. J R Coll Surg Edinb 1999;44:16-8. [22] Wilkiemeyer M, Pappas TN, Giobbie-Hurder A, et al. Does resident post graduate year influence the outcomes of inguinal hernia repair? Ann Surg 2005;241:879-82 [discussion 882-874]. [23] Bickel A, Rappaport A, Hazani E, et al. Laparoscopic cholecystectomy for acute cholecystitis performed by residents in surgery: a risk factor for conversion to open laparotomy? J Laparoendosc Adv Surg Tech A 1998;8:137-41. [24] Bottsford JH, Blake R, Small T, et al. Carotid endarterectomy: the outcome is not adversely affected by surgical residents. J S C Med Assoc 1997;93:243-7.